key: cord-0041885-79ql1z1d
authors: nan
title: Poster Session Abstracts
date: 2009-09-24
journal: Pediatr Pulmonol
DOI: 10.1002/ppul.21133
sha: 8f1e2051af3e525ed4f2b712bdf606d3d0f221b0
doc_id: 41885
cord_uid: 79ql1z1d

nan

The two nucleotide binding domains (NBDs) and the R region of the cystic fibrosis transmembrane conductance regulator (CFTR) are responsible for regulating the chloride pore formed by the membrane spanning domains (MSDs) of CFTR. Deletion of F508 (F508del), a residue within the first NBD (NBD1), is the most common cystic fibrosis causing mutation. F508del leads to reduced cell surface CFTR due to folding and stability defects and a reduction in individual channel activity. F508del causes minimal structural perturbation as demonstrated by comparing crystal structures of WT and mutant NBD1. Instead, the deleterious effects of the mutation are thought to be due to disruption of NBD1 surfaces required for interdomain interactions and/or a change in the kinetics or thermodynamics of NBD1 folding and interactions. We have used NMR to study human NBD1 in order to identify differences in the dynamics of the WT and F508del. Spectra of both WT and F508del hNBD1 with the regulatory insertion removed (deltaRI) exhibit spectral characteristics associated with highly dynamic proteins. The dynamics appear widespread with both the ATP-binding core and the alpha-helical subdomain undergoing motions in the µs-ms timescale range as evidenced by broadening and loss of intensity of NMR resonances from atoms in these regions. The broadening and loss of signal intensity are exacerbated by deletion of F508 with the largest changes occuring at the interface between the ATP-binding core and the alpha-helical subdomain. These results provide evidence that deletion of F508 leads to changes in NBD1 dynamics at sites that are distant from the site of mutation and thus may disrupt the kinetics and thermodynamics of NBD1 interactions even when the interfaces involved are distant from the site of mutation. (Supported by a grant from the CF Foundation.)

Phosphorylation was detected at ten sites, regardless of whether the sample was PKA treated or not. Of these ten sites, nine were in the R region, and one in the neighbouring NBD1 domain. The sites detected with phosphorylation included: S422, S660, S670, S700, S712, S737, S753, S768, S795 and S813. Residues S422 and S670 had never before been detected as phosphorylated in the context of the full length protein. Notably, phosphorylation was detected at the two monobasic PKA consensus sites, S670 and S753, which are considered to be less susceptible to PKA phosphorylation as compared to dibasic sites. In order to establish if PKA treatment induces a change in phosphorylation levels at the ten sites, we designed a MRM method to quantify the abundance of the phosphopeptides containing each of the sites. At each site, there was a significant increase in phosphorylation. Of particular interest is the increase in phosphorylation at the poorly characterized monobasic PKA consensus sites S670 and S753. The increase in phosphorylation at these sites was measured as 26 and 500 fold increases in phosphorylation, respectively. Functional consequences of mutagenesis at these sites will be evaluated to determine their role in channel gating. Studies supported by CCFF and CFFT.

LeSimple, P. 1 ; Robert, R. 1 ; Liao, J. 1 ; Gruenert, D.C. 2, 3 ; Hanrahan, J.W. 1, 4 1. Physiology, McGill Univ., Montreal, QC, Canada; 2. California Pacific Medical Center Res. Inst., San Francisco, CA, USA; 3. Lab. Med., UCSF, San Francisco, CA, USA; 4. McGill Univ. Health Centre Res. Inst., Montreal, QC, Canada The Cystic Fibrosis Transmembrane conductance Regulator (CFTR) is an integral membrane glycoprotein that functions as an anion channel and influences diverse cellular processes. Its role in salt transport has been well studied, but its possible role in maintaining epithelial polarity and barrier function have not been investigated. We studied the effect of expressing wild-type (WT-) or ∆F508-CFTR using the CFBE41o-cell line from human CF bronchial epithelium. Cells were cultured at the air/liquid interface and GFP-tagged WT-or ∆F508-CFTR was transiently expressed using adenoviruses. The results were compared with those obtained from CFBE41olines stably complemented with WT-CFTR (CFBE-WT) or ∆F508-CFTR (CFBE-∆F). CFTR was localized by immunostaining and GFP fluorescence, and barrier function was monitored as transepithelial resistance (TER) and 14 C-mannitol backflux. Although CFTR is an anion channel and its expression would be expected to reduce TER, transient overexpression of GFP-CFTR increased the TER of CFBE41o-monolayers from 380±25 to 638±62 Ω.cm 2 (n=8, p<0.005) whereas transient overexpression of GFP-∆F508-CFTR had no effect (348±47 Ω.cm2). Imaging confirmed that changes in TER were correlated with the localization of CFTR; GFP-CFTR was detected at the apical membrane whereas only perinuclear GFP-∆F508-CFTR was observed. CFBE-WT cells formed high-TER monolayers (1560±94 Ω.cm 2 ) whereas CFBE-∆F had dramatically lower TER= 323±28 Ω.cm 2 . Reducing the temperature to 29∞C for 24h to enhance ∆F508-CFTR processing increased the TER of CFBE-∆F monolayers to 640±108 Ω.cm 2 (n=12, p<0.001), but had little effect on CFBE-WT monolayers, which already had higher TER (826±78 vs control: 868±110 Ω.cm 2 ). Rescue at 29∞C reduced 14 C-mannitol backflux and increased the TER of CFBE41ocell monolayers to values near those observed for CFBE-WT monolayers (1005±85 vs 494±81 Ω.cm 2 at 37∞C, n=8, p<0.001), and the effect of temperature on TER was reversible through two cycles of restrictive and permissive temperatures. The results suggest that CFTR expression at the plasma membrane strongly influences TER, and this effect is probably at the level of tight junctions because 14 C-mannitol is an indicator of paracellular permeability and the changes in TER are opposite to those expected for insertion of a conductance in the membrane. The temperature dependence of TER was not dependent on the Clconcentration and was not affected by CFTRinh172, which strongly inhibited forskolin-stimulated current, suggesting that CFTR-induced changes in barrier function do not depend on its channel activity. After ∆F508-CFTR rescue by low-temperature, genistein (10 µM) reduced the TER to the value observed without rescue (888±152 vs 561±96 Ω.cm 2 ; n=7, p< 0.05, vs 37∞C: 580±93 Ω.cm 2 ), suggesting a possible involvement of a tyrosine kinase. CFTR trafficking to the apical membrane may be required for the optimal organization of junctional complexes, and partial loss of this barrier function in tissues that express ∆F508-CFTR may contribute to epithelial abnormalities in CF. Support: CCFF and CIHR (Breathe Program), CFFT, CFF, MUHC and Pennsylvania CF, Inc.

Wright, J.M. 1 ; Joseloff, E. 2 ; Nikolsky, Y. 3 ; Serebriyskaya, T. 3 ; Wetmore, D. 2 1. Physiology, Johns Hopkins Medical Institutions, Baltimore, MD, USA; 2. Cystic Fibrosis Foundation Therapeutics, Bethesda, MD, USA; 3. GeneGo, Encinitas, CA, USA One unresolved issue in cystic fibrosis research is whether the defective CFTR trafficking and resulting functional loss is directly linked to the often seen chronic inflammatory state, and if so what is the possible mechanism. OMICs experiments investigating protein or gene expression changes due to altered CFTR trafficking have produced long lists of elements significantly changed in expression but have no apparent connection to inflammation. Likewise, microarray experiments documenting the effects of Pseudomonas aeruginosa on bronchial epithelial cell gene expression patterns have yielded no insights into CFTR trafficking or its role in the infection process. In an attempt to understand the possible interplay between inflammation and CFTR trafficking, we combined and analyzed the results of several published OMIC studies using MetaMiner (Cystic Fibrosis), a knowledge base data analysis system created by CFFT and GeneGo, Inc. The results of 7 OMICs studies were entered into the analysis system and examined for known, published interactions between the experiments. Unexpectedly, numerous connections were established between genes documented to correct DF508 trafficking when expressed in cell lines (Trzcinska-Danelut, et. al. 2009 ) and a list of genes differentially expressed in bronchial epithelial cells after exposure to Pseudomonas aeruginosa (Mayer et.al. 2007 GEO dataset GSE6802) . Of 34 genes documented to correct DF508 trafficking, 9 (CCL2, Endothelin-1, Calgranulin A, CCK, AQP3, Galectin-3, Caveolin-2, STAT1, PPAR-gamma) were directly linked by positive expression activation mechanisms to the inflammatory response. Two of the nine correctors, CCL2 and AQP3, were increased in expression in the bacterial infection model. STAT1, a trafficking corrector, is known to promote HSP90 and HSP70 expression which are involved in CFTR trafficking. However, STAT1 also influences other inflammatory response transcription factors which in turn activate expression of CCL2, Endothelin-1, CCK and Galectin-3. Looking at interactions among the results as a whole and in detail, it appears that an inflammatory response produces numerous changes which favor correct trafficking of DF508. In that regard, one can take an alternative view of the inflammatory process as potentially a compensatory mechanism for dysfunctional DF508 trafficking. Using knowledge management platforms to integrate multiple disciplines within the CF research community should accelerate our understanding of disease mechanism and identify therapeutic targets.

Supported by Cystic Fibrosis Foundation Therapeutics.

repair the function of defective CFTR. VX-770 is an orally bioavailable CFTR potentiator in clinical development for the treatment of cystic fibrosis. We previously reported that VX-770 potentiates CFTR-mediated chloride secretion in vitro using both recombinant cell lines and human bronchial epithelium (HBE) cultured from the bronchi of donor lung tissue.

In the current study we evaluated the effect of VX-770 on Na + transport, fluid transport, and cilia beating in HBE carrying the G551D CFTR gating mutation on one allele and the F508del processing mutation on the other allele (G551D/F508del-HBE). In Ussing chamber experiments, VX-770 plus forskolin increased CFTR-mediated Clsecretion in G551D/F508del-HBE by approximately 10-fold to about 50% of that observed in HBE isolated from individuals without cystic fibrosis (EC 50 = 236 ± 200 nM). The increase in CFTR function by VX-770 was sufficient to reduce the excessive epithelial Na + channel (ENaC)-mediated Na + absorption to levels observed in wild-type HBE as determined by the change in potential difference response to amiloride (-46 ± 2 mV in untreated G551D/F508del-HBE vs. -22 ± 1 mV in VX-770-treated G551D/F508del-HBE vs. -17 ± 4 mV in wildtype-HBE). VX-770 had no effect on ENaC function in recombinant cells, suggesting that the decrease in Na+ transport was the result of CFTR potentiation. We further found that VX-770 was able to decrease the airway surface liquid absorption rate in G551D/F508del-HBE towards that observed in wild-type HBE (1.25 ± 0.06 µl/cm 2 /hr in untreated G551D/F508del-HBE vs. 1.0 ± 0.03 µl/cm 2 /hr in VX-770-treated G551D/F508del-HBE vs. 0.82 ± 0.01 µl/cm 2 /hr in wild-type HBE), which is consistent with the decrease in ENaC-mediated Na + transport. Finally, the increase in airway surface volume caused by VX-770 was sufficient to increase the cilia beating frequency from 2.9 ± 0.3 to 11 ± 1 Hz in G551D/F508del-HBE. These results support the hypothesis that pharmacological agents that increase defective CFTR function may rescue ion and fluid transport in cultured human cystic fibrosis airway cells. We have identified and validated a variety of chemically and structurally distinct molecules as correctors of the CFTR-∆F508 trafficking mutation. The next stage in the drug discovery process will be discovery of mechanism(s) of action leading to correction. To elucidate the events involved in correcting CFTR-∆F508 we generated a compendium of genome-wide corrector transcriptional profiles.

Our genomics pipeline comprises data generated from treating HEK293 cells and a more relevant lung cell line, polarized CFBE41ocells (human airway cells derived from a patient homozygous at ∆F508), with correctors and then transcriptionally profiling the cellular response to correction.

We chose the Agilent platform, and labeled each RNA sample with Cy3 and Cy5. Each sample is hybridized onto an expression array containing 44,000 probes to the human genome, with a universal reference RNA labeled with the reciprocal dye. The data is further normalized against the vehicle treated control and a biological duplicate was done for each treatment. We have generated temporal transcriptional profiles for correction by treating cells for 1hr, 6hr, 12 hr and 24 hrs.

As proof of principle we queried specific genes known to be associated with CFTR folding and trafficking and found they were significantly (p∠0.05) and differentially expressed. For instance, the chaperone Hsp70 reported to associate preferentially with the mutant CFTR-∆F508, was significantly down-regulated across several drug treatments in the HEK293 cell line, and RAB11A is significantly down-regulated in the polarized airway cells across a large number of corrector signatures. We have derived distinct transcriptional profiles for our correctors and clustering is apparent in categories relating to CFTR trafficking.

So far, we have identified a novel gene, involved in N-glycan biosynthesis which is significantly down-regulated upon treatment with one of our correctors, a sodium pump inhibitor, Ouabain. Targeting this gene with siRNA in human epithelial cells expressing CFTR-∆F508 restored expression at the cell surface by 2 fold over control, as determined by flow cytometry. We are in the process of validating other hits generated from this data by siRNA and overexpression, real-time PCR and Western blotting. Generating such profiles will elucidate novel pathways and mechanisms of CFTR rescue, and drive a new generation of target-driven therapies.

cell line (CFBE41o-). A total of 354 hits were identified and confirmed in the primary assay.

Validation of the hits included a lack of cytotoxicity of the shRNA, expression of target in airway epithelial cells, identification of a second shRNA against the same target, and cell surface expression by biotinylation of CFTR ∆F508 upon shRNA-mediated knock-down of the target. Most importantly, these shRNAs restored functional activity of the mutant channel in primary bronchial epithelial cells from ∆F508/∆F508 CF patients in transepithelial CFTR mediated current assays (Table 1 ). There was a very strong correlation between fully glycosylated band C expression in the CFBE41o-cells and functional activity in the primary cells of CF patients.

We will present an overview of the target discovery and validation program, resulting in a portfolio of 19 novel drug targets to treat CF. The identity of these novel targets also sheds light on the biology of trafficking of CFTR. For example, these data point to a role of TGF-beta signaling and inflammatory mediators in airways in the regulation of CFTR trafficking.

Acknowledgments: We thank Drs. Bill Guggino, Ineke Braakman, Kevin Foskett, John Hanrahan and Hugo De Jonge for helpful discussions. We acknowledge the support of Cystic Fibrosis Foundation Therapeutics. The most common cystic fibrosis (CF)-causing mutation in the cystic fibrosis transmembrane conductance regulator (CFTR) is deletion of Phe508 (∆F508) in the first of two nucleotide binding domains (NBDs). The ∆F508 mutation leads to defects in cellular processing and folding, as well as channel gating. As a potential therapy for CF, small-molecule compounds have been identified that correct processing/folding defects (correctors) and/or channel gating defects (potentiators). We report NMR studies of the interactions of a number of newly identified small-molecule compounds with wild type (WT) and ∆F508 human NBD1. In order to obtain appropriate solubility and yields required for NMR, our constructs lack the regulatory insert (∆RI) normally contained within NBD1. These constructs either retain or omit the regulatory extension (∆RE), which is composed of the first 25 residues of the R region. Addition of several related compounds to both WT and ∆F508 (∆RI ∆RE) (387-646) gives NBD1 spectra with changes in chemical shifts for residues in the two short C-terminal helices and intervening loop (634) (635) (636) (639) (640) (641) (642) (643) (644) (645) (646) . In contrast, minimal shifts were observed when compounds were added to constructs retaining the RE in . We have previously reported that phosphorylation disrupts interactions between the murine NBD1 core and the RI/RE regions. Titration of compounds into phosphorylated constructs for both WT and ∆F508 reveals similar chemical shifts although to a lesser degree. We interpret these results to imply that the small-molecule compounds compete with the C-terminal helix and the RE/R region for interaction with the NBD1 core. We hypothesize that binding to the C-terminal site displaces the RE/R region from its NBD1 site, thereby promoting NBD dimerization required for ATP hydrolysis and accounting for the potentiating effects of the compounds. Additionally, we suggest that promoting dimerization could stabilize the folding of full-length CFTR, accounting for the corrector effect of the compounds. This work was funded by the CF Foundation and EPIX Pharmaceuticals.

Despite great efforts to elucidate the mechanism and the molecular factors involved in CFTR biogenesis, trafficking and function, many of these processes are not fully understood. Protein kinases and phosphatases have long been known to regulate CFTR function. However, the role of phosphorylation in CFTR biogenesis and trafficking remains less explored. Recently, it was shown that recombinant casein kinase 2 (CK2) phosphorylates in vitro CFTR NBD1 at the residue serine 422 [1] . In addition, CFTR possesses several CK2 consensus phosphorylation sites, namely S511 (very close to F508) and T1471 at the C-terminus. Our aim here was to determine how mutation of these putative CK2 P-sites affects the CFTR biogenesis, turnover and processing.

We have produced CFTR mutants in which the consensus residues S422, S511 and T1471 were substituted by either a neutral (alanine, A) or an acidic residue (aspartic acid, D) in both wt and F508del-CFTR backgrounds by site-directed mutagenesis and used these constructs to generate novel stable BHK cell lines. Pulse-chase experiments followed by CFTR immunoprecipitation, Western blot assays and functional assessment by iodide efflux assay were performed in these cells. Quantification of bands B (immature form) and C (mature form) of CFTR in pulse-chase shows that substitution of S511 does not affect the turnover or processing of either wtor F508del-CFTR. However, substitution of threonine at position T1471 to D (but not to A) completely impairs the processing of wt-CFTR and increases the turnover of F508del-CFTR. Interestingly, T1471A although not interfering in the production of mature protein, abolishes CFTR function. Mutation of S422 to an aspartic acid residue, although not compromising the appearance of mature CFTR, reduces its function by 40%. Moreover, treatment of cells with 20 µM TBB (tetrabromobenzotriazole, a specific inhibitor of CK2) for 90 min significantly reduces the processing efficiency of wt-CFTR.

Our data suggest a putative stabilizing role for CK2 upon wt-CFTR in these cells, which can be mediated by CFTR residues S422 or T1471. So far, our data exclude any role for residue 511 on the functional interaction of CFTR and CK2.

Work 

The activity of cytoplasmic domains of CFTR controls channel opening and closing of the anion-selective pore. The channel activation requires phosphorylation of the R-domain while ATP interacts at the two nucleotidebinding domains (NBD), playing a critical role on the channel gating. We discovered several years ago the dual activity compounds benzo [c] quinolizinium and hypothesized that the NBD1-R domain tandem could be an important molecular site on CFTR for interaction with these compounds (Becq et al. J Biol Chem 1999; 274:27415; Dormer et al. J Cell Sci 2001; 114:4073; Stratford et al. Biochem Biophys Res Commun 2003; 300:524; Norez et al. J Pharmacol Exp Ther 2008; 325:89) . We also performed a molecular dissection of this part of CFTR by examining several CFTR mutants (constructed by site-directed mutagenesis and expressed as GFP-tagged proteins in HEK and BHK cells) situated in the neighborhood of the putative latest β strands (β c 5 and β c 6) of the NBD1 (Callebaut et al. Cell Mol Life Sci 2004; 61:230) . Two glycine residues appear highly conserved in ABC transporters and are suspected to form the hairpin (G622) between the two β strands or to be located at the extremity of the last β S-Nitrosoglutathione (GSNO) is an endogenous signaling molecule with many potentially beneficial effects. GSNO is normally present in the lung but concentrations tend to be low in cystic fibrosis (CF) patients. Previously, we and several other research groups have found that different Snitrosylating agents, including GSNO, increase the expression, maturation and function of mutant ∆F508 CFTR in human airway epithelial cells. Further, we showed that ∆F508 CFTR maturation leads to the cell surface expression following GSNO treatment. We also examined the relative ability of GSNO and hypothermia (27 o C) to up-regulate ∆F508 CFTR expression. We have also demonstrated a role for different molecular co-chaperones, including Hsp70/Hsp90 organizing protein (Hop), in the regulation of mutant ∆F508 CFTR expression. Also of interest, we found through Western blot analysis and more recently by immunofluorescence and confocal microscopy that Hop is expressed in CFBE41o-cells, CFPAC-1 cells and A549 cells. In addition, we showed that GSNO S-nitrosylated Hop decreases Hop expression and also decreases the association between Hop and ∆F508 CFTR in the ER. Interestingly, knockdown of endogenous Hop by using siRNA duplexes for Hop, produced marked increases in levels of fully mature forms of ∆F508 CFTR; but, in the presence of GSNO, expression of mature form of ∆F508 CFTR, significantly increased. At present, we sought to determine whether GSNO and hypothermia increase stability and decrease the internalization rate of mutant ∆F508 CFTR. CFBE41o-cells were grown at 37 o C to 70% confluence, and then incubated for an additional 48 hr at 27 o C in the presence or absence of GSNO (10 µM) for last 4 hr. Internalization rates were then performed at 37 o C, using a two-step biotinylation processes. Following biotin labeling cells were lysed and CFTR was immunoprecipitated with an anti-CFTR antibody (kindly provided by Dr. J. Riordan). Interestingly, we found that both GSNO and hypothermia not only induce mutant ∆F508 CFTR expression but also enhance stability of ∆F508 CFTR in the cell surface and decrease the internalization rate of ∆F508 CFTR in CFBE41o-cells. On the other hand, the combination of both treatments (GSNO/hypothermia) has a greater effect on stability and internalization rate of ∆F508 CFTR than either alone. These data indicate that GSNO treatment may be complemented by other corrector therapies to exploit more than one mechanistic pathway, achieving the restoration of mutant ∆F508 CFTR function. Supported by the CF Foundation and the NIH.

Luz, S.F. 1, 2 ; Romeiras, F.M. 1 ; Matos, P. 2 ; Mendes, A. 2 ; Jordan, P. 2 ; Amaral, M.D. 1, 2 ; Farinha, C.M. 1, 2 1. Chemestry and Biochemestry Department, Faculty of Sciences University of Lisboa, Lisboa, Portugal; 2. Human Genetics Department, National Institue of Health Dr. Ricardo Jorge, Lisboa, Portugal Regulation of CFTR early and late intracellular trafficking as well as channel activation is the result of a complex network of CFTR interacting proteins (CIPs) namely, molecular chaperones, glycosidases, the basal trafficking machinery (Rab GTPases, SNAREs and PDZ-domain-proteins) and other factors including molecular switches (like protein kinases and phosphatases). Among CIPs yet to be identified and/or characterized are those affecting CFTR biogenesis/traffic by phosphorylation and dephosphorylation. Spleen tyrosine kinase (SYK) is a non-receptor tyrosine kinase, described to have a role in signal transduction and signalling in haematopoietic cells. Its consensus for protein phosphorylation includes a tyrosine followed by two negative residues. In CFTR, this consensus appears only once at residue 512 (i.e., very close to F508, the residue which is deleted in most CF patients).

Our aim here was to identify the putative role of SYK upon CFTR biogenesis and trafficking. To this end, we produced cell lines stably expressing CFTR mutants, where Y512 was substituted by either a neutral residue (alanine, A) or by an acidic residue (aspartic acid, D) in both the wt-and F508del-CFTR backgrounds. Pulse-chase experiments followed by CFTR immunoprecipitation and Western blot were performed using these cells lines. Quantification of bands B (immature form) and C (mature form) of CFTR shows that the substitution of Y512 by either A or D decreases both the steady-state levels and the efficiency of processing of CFTR. The iodide efflux assay was used to functionally assess these CFTR variants and the results show that cells expressing the Y512A-CFTR variant have 15% less IBMX/Fsk-stimulated iodide efflux than wt-CFTR, whereas Y512D-CFTR shows a 60% reduction.

Moreover, in vitro phosphorylation assays, using immunoprecipitated SYK and purified CFTR NBD1, show that in vitro SYK not only undergoes autophosphorylation but also phosphorylates CFTR NBD1, this effect being abolished when a dead kinase mutant of SYK is used in the same assay. RT-PCR analysis of epithelial respiratory cells (obtained by nasal brushing from either non-carrier controls and F508del homozygous patients and also the established cell lines Calu-3 and virally transduced wt-or F508del-CFBE) show that SYK is endogenously expressed in these cells, arguing for the possible physiological significance of these findings.

Altogether, our data suggest a putative positive role for SYK on wt-CFTR stability and processing and that this effect seems to be mediated by the CFTR consensus site for SYK at Y512. At present, indirect effects (i.e., independent of direct CFTR phosphorylation by SYK) cannot be completely put aside.

Work supported by POCTI/PTDC/BIA-BCM/67058/2006 grant, EU grant FP6-LSH-2005-037365 (TargetScreen2) and pluriannual funding of BioFig (FCT, Portugal).

High-throughput screenings for correctors of ∆F508 CFTR protein processing have yielded potential candidate compounds, but the pharmacological utility of these reagents requires assessment of correction in physiological testing systems. Western blot analysis of human ∆F508 CFTR expressed in C127 cells and measurements of CFTR-mediated I sc across gallbladder mucosa from ∆F508 CFTR mice (which have normal levels of CFTR transcription, cftr tm1Eur , gift of Dr. B.J. Scholte, Erasmus University) were employed in the evaluation of publicly available correctors from the CFF Modulators Program. Western blot analysis of ∆F508 CFTR expressing C127 cells were treated with 10 µM C3-C6; an increase in band C CFTR was only detected for C6 treatment. To assess correction of function, aseptic gallbladder organ preparations were mounted in polycarbonate cups and maintained in DMEM/F12 + pen/strep media at 37∞C in a 95% O 2 : 5% CO 2 atmosphere to avoid CFTR down-regulation by hypoxia. The gallbladder preparations were treated on both apical and basolateral sides with either DMSO vehicle or corrector (C) concentrations equivalent to the reported K a (1-3 µM) or 10 µM for 24 hrs. After treatment, the gallbladder cultures were mounted in Ussing chambers in standard Kreb's bicarbonate Ringers containing the same concentration of corrector. After equilibration, gallbladder preparations were treated with a forskolin/IBMX/genistein cocktail for 15 min followed by sequential treatments with 25 µM CFTR(inh)-172 and 100 µM niflumic acid (NFA) for 10 min each. Vehicle control studies showed nearly complete loss of the CFTR(inh)-172-sensitive I sc (CFTR ∆I sc ) and small increase in the NFA-sensitive I sc , a measure of alternative Clconductances (ACC ∆I sc ), in the ∆F508 CFTR as compared to wild-type preparations (WT: CFTR ∆I sc = 62.1±9.0 and ACC I sc = 2.1±1.1 vs. ∆F: CFTR ∆I sc = 1.6±0.9 and ACC I sc = 5.6±0.5 µA/cm 2 , p<0.05, n = 4 -8) . The corrector C1 demonstrated a positive concentration-response relationship for the CFTR ∆I sc (3 µM = 3.2 ± 0.0; 10 µM 6.4 ± 1.3) but not for the ACC ∆I sc (3 µM = 7.4 ± 2.8; 10 µM 3.2 ± 2.6, n = 3-4). The corrector C3 did not correct the CFTR ∆I sc at 2 µM but increased both CFTR ∆I sc and ACC ∆I sc at 10 µM (CFTR ∆I sc : = 11.1±8.0; ACC ∆I sc = 8.0±1.6, n =2). Measurements of the transepithelial conductance under all conditions did not indicate adverse effects of C1 or C3 on mucosal integrity. Evaluation of other correctors is on-going. In conclusion, positive effects of the publicly available correctors C1, C3 and C6 on ∆F508 CFTR processing were found using Western blot analysis and/or native murine gallbladder mucosa. However, the functional studies indicate that the level of correction (10-18%) at the concentrations used is likely near the lower limit of therapeutic value for cystic fibrosis patients.

Supported by NIH, CFFT, and EUROCARE CF. A number of small molecules have been reported to modulate CFTR maturation and function. These compounds can be classified according to their effects on CFTR: small molecules that modulate CFTR function are termed potentiators or inhibitors, whereas small molecules that rescue mutant CFTR maturation efficiency are termed correctors. One possible mechanism by which these compounds may affect CFTR, is by direct interaction. We have developed assays to test whether a given compound is able to directly interact with the critical NBD1-CFTR or predicted sites within full length CFTR. NBD1-CFTR contains the most common CF-causing maturation mutation, the deletion of phenylalanine at position 508, a mutation that destabilizes the isolated domain; known NBD1-CFTR ligands (such as ATP) stabilize the domain. The panel of CFFT compounds and other known CFTR modulators were assessed for direct interaction with NBD1-CFTR utilizing a stabilization assay. Under mild denaturant conditions, a number of the tested compounds were able to protect the native conformation of NBD1-CFTR by directly interacting with this subdomain of CFTR. In addition, specific, predicted, interdomain binding sites shared with NBD1-CFTR were tested with mutants designed to fill these sites in full length CFTR. Identification of the site(s) of action of small molecule CFTR modulators provides critical mechanistic information and is the first step towards elucidating relevant structures.

(Supported by CFF and NIH NIDDK to PJT.)

Cystic fibrosis (CF) is a fatal disease affecting the lungs and digestive system caused by mutations in the gene coding for the cystic fibrosis transmembrane conductance regulator (CFTR) protein. Many aspects of both wild type (WT) and mutant CFTR regulation of transcription, maturation, trafficking and degradation are still unclear.

The SUMO protein is covalently conjugated to a lysine residue embedded in a specific target motif Ψ-K-X-E and thereby regulates a large number of mostly nuclear proteins. SUMOylation is a reaction involving different E-ligases. The key enzyme is UBC9, binding SUMO and the target protein for the final covalent binding of SUMO. SUMOylation is also a highly dynamic process, because SUMO specific proteases (SENP) quickly de-SUMOylate the target protein. As SUMOylation may also play a role for proteins targeted to the plasma membrane, we hypothezised the regulation of CFTR trafficking by SUMOylation.

We identified 2 highly conserved SUMO binding sites in the CFTR protein and demonstrated the binding of SUMO conjugation enzyme UBC9 to both WT and ∆F508 CFTR. Using directed yeast two-hybrid tests and coimmunoprecipitation analysis we showed SUMOylation of CFTR at the 2 target sites, both in vitro and in vivo.

Elimination of the SUMO sites by site-directed mutagenesis or overexpression of the SUMO specific protease SENP1 resulted in a substantial decrease of fully glycosylated WT CFTR, whereas the abundance of ∆F508 CFTR increased. Cell surface biotinylation experiments confirmed these findings and mutations of both SUMO motifs in WT CFTR inhibited its membrane transport.

Gradient centrifugation experiments revealed that WT CFTR with mutations in the SUMO motif can escape the ER but is unable to leave the Golgi apparatus.

In summary, we present a new mechanism for the regulation of CFTR maturation and membrane trafficking. This work is funded by Deutsche Forschungsgemeinschaft (DFG).

Ehrenfeld, J.; Loriol, C.; Avella, M.; Bouloukos, K.; Dulong, S.; Borgese, F. FRE3094, CNRS/UNSA, Nice, France SLC26A9 behaves as a Clchannel preferentially expressed in lung and gastric epithelia. By analogy with other SLC26 family members, we investigated possible functional-and protein-interactions between SLC26A9 and CFTR, in two protein expression systems. We had previously cloned SLC26A9 from primary culture of bronchial cells, NHBE (Loriol et al. Cell Physiol Biochem 2008; 22:15) .

In that aim, bronchial cell lines expressing wt-CFTR (CFBE41o-wtCFTR) or ∆F508-CFTR (CFBE41o--∆F508-CFTR) or not expressing CFTR (CFBE41o-) were transduced with SLC26A9. The expression of the protein was controlled by Western blot. Anion transport short circuit current (SCC) was measured by applying a serosal to mucosal directed Clgradient through the reconstituted epithelium, followed by forskolin or forskolin/genistein application.

Larger currents (two times) were found in CFBE41o-cells transduced with SLC26A9 compared to control cells; 10 µM niflumic acid poorly affected SCC in SLC26A9 transduced cells and was ineffective in control cells.

In CFBE41o-wtCFTR, forskolin-stimulated currents were 97 ± 33% larger (n=15; p<0,05) in cells transduced with SLC26A9 compared to control cells; forskolin-stimulated currents were blocked in both batches of cells with the same percent of inhibition by Treated 48 h at 27∞C, CFBE41o-cells expressing ∆F508-CFTR and transduced with SLC26A9 also presented larger currents (two times) than control cells when a serosal to mucosal directed Clgradient was applied.

However, forskolin/genistein-stimulated currents and level of CFTR-inh172 inhibition were not significantly different in the two batches of cells. At 37∞C, CFBE41o-cells expressing ∆F508-CFTR and transduced with SLC26A9 did not respond to forskolin/genistein.

Xenopus laevis oocytes co-injected with SLC26A9 and wt-CFTR presented larger currents upon forskolin stimulation than the sum of currents measured in oocytes expressing SLC26A9 or wt-CFTR alone. The implication of CFTR, in the stimulatory effect obtained in oocytes co-expressing SLC26A9 and CFTR, was deduced from the % of inhibition (more than 80%) of forskolin-stimulated currents by application of CFTRinh-172 (10µM) plus glybenclamide (200µM). This effect was not associated to SLC26A9 since CFTR blockers had a maximal 20% inhibitory effect in oocytes expressing SLC26A9 alone.

Co-immunoprecipitation of SLC26A9 and CFTR was found in oocyte membranes and in bronchial cells transduced with both anion channels.

In conclusion, while SLC26A9-associated currents were clearly detected in oocytes, the absence of a specific pharmacology was a limit to strictly associate SLC26A9 to the observed increased currents in CFBE41o-cells transduced with SLC26A9 alone. However, in both amphibian and mammalian expression systems, SLC26A9 stimulates the functioning of wt-CFTR. In addition, the co-immunoprecipitation studies point to a physical interaction between both anion channels in agreement with other SLC26 family members. The functional benefit of co-expression is not seen with ∆F508-CFTR and SLC26A9.

Acknowledgements: Supported by "Vaincre la Mucoviscidose," CNRS, UNSA-France. Previous work has shown that CFTR-NBDs possess the capacity for two different kinds of enzymatic activity that are coupled to channel opening and closing. In the presence of ATP alone, ATPase activity (ATP Ç ADP + P i ) governs normal channel gating. However, when ATP is added in the presence of adenosine 5'-monophosphate (AMP), adenylate kinase activity (ATP + AMP Ä ADP + ADP) regulates channel activity. Adenylate kinases have distinct binding-sites for ATP and AMP that often bind cooperatively. Although structural and functional studies have ascertained where ATP binds in the NBDs of CFTR and other ATP-binding cassette (ABC) transporters, the amino acid residues involved in AMP-binding and the mechanisms underlying adenylate kinase activity and adenylate kinase-dependent CFTR gating are unknown. We hypothesized that as in other adenylate kinases, AMP binds to a site in CFTR that is distinct from the ATP-sites and that ATP and AMP mutually influence their interaction with CFTR.

To test this hypothesis we studied the interactions of the 8-azido-photoreactive analogues of ATP, AMP and P 1 ,P 4 -di(adenosine-5') tetraphosphate (Ap 4 A) with membrane-bound CFTR. As others have reported, we found that [α 32 P]8-N 3 -ATP labeled CFTR. Non-radioactive ATP reduced labeling, suggesting specific labeling of the ATP-sites. AMP did not reduce 8-N 3 -ATP labeling indicating that it did not compete at the ATP-sites. Instead, AMP increased 8-N 3 -ATP labeling by ~20% suggesting that it induced conformational changes that involve the ATP-sites. When we added [α 32 P]8-N 3 -AMP alone, it only weakly labeled CFTR. However, when we repeated the experiment in the presence of ATP, labeling with [α 32 P]8-N 3 -AMP increased several fold. This labeling decreased in the presence of nonradioactive AMP suggesting specific labeling of the AMP-site. 8-N 3 -AMP labeling of CFTR also decreased when Ap 4 A or P 1 ,P 5 -di(adenosine-5') pentaphosphate (Ap 5 A) was added; Ap 4 A and Ap 5 A are compounds that bind both ATP-and AMP-sites of adenylate kinases. In addition, we found that CFTR was readily labeled with [β 3 32 P]8-N 3 -Ap 4 A and that labeling was reduced when we added non-radioactive Ap 4 A or Ap 5 A. Importantly, photolabeling was reduced both by non-radioactive ATP and by AMP, consistent with [β 3 32 P]8-N 3 -Ap 4 A interacting with both ATP-and AMP-sites in CFTR. In summary, these results indicate that 1) AMP binds to a site within CFTR that is different from the ATP-sites; 2) ATP enhances specific labeling of the AMP-site; 3) AMP induces conformational changes that involve the ATP-sites; and 4) ATP and AMP interactions with CFTR show some similarities to other adenylate kinases. Specific labeling of the AMP-site could provide a tool to identify labeled CFTR peptides that are involved in AMP-binding.

Supported by CFF, Francis Family Foundation, and HHMI. 

Tukaye, D.N.; Guggino, W.B. Johns Hopkins University, Baltimore, MD, USA Type III Secretion System (T3SS) toxins of P. aeruginosa are important virulence factors in Pseudomonas infections. In a mouse model of Pseudomonas pneumonia, infection with P. aeruginosa Exo S+ strains causes an increase in total amount of fluid in the lungs as determined at autopsy in contrast to infection with P. aeruginosa Exo Sstrains. The exact molecular mechanism underlying these observations is not known. ExoS is a bifunctional protein with GTPase Activating Protein (GAP) activity at the N terminal domain and ADP ribosyl transferase (ADPRT) activity at the C terminal domain. We find that ExoS-GAP activity increases total cellular levels of mature (C band) Wild Type (WT) CFTR in CFBE41o-and MDCK cells as measured by Western blot analysis. R146K, a mutant of ExoS-GAP, lacking the GAP activity fails to significantly increase WT CFTR total levels. This indicates that GAP activity conferred by the R146 residue is essential for its effect. In contrast, ExoS-GAP fails to increase total levels of band B of ∆508 CFTR. This indicates that targets of ExoS-GAP exist in CFTR trafficking beyond the ER degradation pathway. ExoS-GAP fails to increase total levels of Na + /K + ATPase indicating that ExoS-GAP targets are mediators of trafficking pathways specific to CFTR and related proteins. Increase in total CFTR by ExoS is mediated by decreased delivery of WT CFTR for lysosomal degradation. Protein turnover studies corroborate this mechanism wherein turnover of WT CFTR in cells transfected with ExoS-GAP is much slower than in control cells or in cells transfected with R146K. ExoS-GAP also brings about a corresponding increase in surface levels of mature WT CFTR.

In conclusion, we show that P. aeruginosa T3SS ExoS-GAP upregulates total and surface levels of mature WT CFTR by modulating CFTR trafficking beyond the ER, most likely by decreasing lysosomal degradation. The decrease could be occurring in part by inactivation of small molecular weight G-proteins involved in delivery of WT CFTR to lysosomes. Increases in total and hence surface levels of CFTR could in part explain increased amounts of fluids seen in the P. aeruginosa pneumonia mouse model treated with Exo S+ strains. This study is supported by CFF and NIH. The balance between protein biosynthesis and degradation is one of the most important determinants for CFTR delivery to, and function at, the cell surface. In the endoplasmic reticulum (ER), this balance is determined by proteins that facilitate folding (i.e. the Sec61 translocon and molecular chaperones) and components of the ubiquitin-proteasome pathway. The heat shock proteins of 70 kDa, constitutive Hsc70 and inducible Hsp70 (referred to here as Hsp70), reside at the junction of these two competing pathways. During CFTR biosynthesis, Hsp70 together with cochaperones (i.e. Hdj1 and Hdj2) is proposed to assist folding and maturation. The resulting improvement in folding efficiency is expected to render CFTR less susceptible to proteasomal degradation. In contrast, Hsp70 is also reported to recruit E3 enzyme complexes (i.e. CHIP-UbcH5a) that facilitate covalent ubiquitin attachment needed for proteasome targeting. The mechanism and extent to which the Hsp70 system contributes to these opposing fates for CFTR remain unknown.

To address this question, we developed a rabbit reticulocyte lysate (RRL)-based cell-free system that enables separation of CFTR synthesis from degradation and investigated the pro-folding and pro-degradative functions of Hsp70. We now selectively manipulate cytosolic Hsp70 function Organellar acidification by the electrogenic vacuolar proton-ATPase is coupled to anion uptake and cation efflux to preserve electroneutrality. The defective organellar pH regulation, caused by impaired counterion conductance of the mutant cystic fibrosis transmembrane conductance regulator (CFTR), remains highly controversial in epithelia and macrophages. Impaired functional expression of CFTR has been implicated in the hyperacidification, and alkalinization, as well as having no effect on organellar pH regulation. Restricting the pH-sensitive, fluorescein-isothiocyanate (FITC) probe to CFTR containing vesicles, the counterion and proton permeability, and the luminal pH of endosomes were measured in various cells, including genetically matched CF and non-CF human respiratory epithelia, as well as cftr+/+ and -/-mouse alveolar macrophages. Passive proton and relative counterion permeabilities, determinants of endosomal, lysosomal and phagosomal pH-regulation, were also probed with FITC-conjugated transferrin, dextran and Pseudomonas aeruginosa, respectively. Although CFTR function could be documented in recycling endosomes and immature phagosomes, neither CFTR channel activation nor inhibition influenced the pH in any of these organelles. CFTR heterologous overexpression also failed to alter endocytic organellar pH, implying that the chloride accumulation does not limit the vesicular acidification. We propose that the relatively large CFTR-independent counterion and small passive proton permeability ensure efficient shunting of the proton-ATPase generated membrane potential. These results have implications in the regulation of organelle acidification in general and demonstrate, consistent with a subset of previous studies, that perturbations of the endo-lysosomal organelles pH homeostasis cannot be linked to the etiology of the CF lung disease. Supported by CIHR and NIH. CFTR∆F508 is the most common mutation in cystic fibrosis. It has at least three defects: it disrupts biosynthetic processing, it reduces the channel opening rate, and it decreases the protein lifetime on the cell membrane. Unlike human CFTR∆F508, mouse CFTR∆F508 is partially processed to the cell membrane, yet it exhibits a gating defect similar to hCFTR∆F508. To better understand which regions of the protein are important in causing CFTR∆F508 processing and gating defects, we designed a series of humanmouse CFTR chimeras including the membrane spanning domain 2 (MSD2) and the two nucleotide binding domains (NBD1 and NBD2). We found that substitution of mNBD1 partially rescued the processing defect of CFTR∆F508; the C-terminal portion of mNBD1 (the Regulatory Extension, RE) was critical for this rescue. The crystal structures of CFTR-NBD1 suggest that the RE region lies on the surface of NBD1 and therefore may be accessible for interaction with other domains in CFTR or with other proteins. In contrast, none of the mNBD1 chimeras rescued the gating defect. Models of CFTR structure suggest interactions between NBD1 and the intracellular loops (ICLs) of MSD2. Therefore, we substituted mMSD2 into hCFTR and found that it did not rescue the CFTR∆F508 processing defect, but it did affect gating. The mMSD2 chimera reduced the channel opening rate of CFTR. However, adding the ∆F508 mutation caused no further decrease in opening rate. We found that both ICL3 and ICL4 of MSD2 are important for this change. These results indicate that the CFTR∆F508 processing and gating defects involve different mechanisms. By pointing to specific regions of CFTR that alter the effects of ∆F508, the data may sug-gest novel strategies for correcting the defects caused by the ∆F508 mutation.

Jiang, H. 1 ; Ramos, A.A. 1 ; Coutermarsh, B.A. 2 ; Stanton, B.A. 2 ; Yao, X. 1 1. Department of Chemistry, University of Connecticut, Storrs, CT, USA; 2. Department of Physiology, Dartmouth Medical School, Hanover, NH, USA The goal of this study was to develop an accurate, precise and high throughput method for quantitative analysis of plasma membrane CFTR. In this work, we report the first mass spectrometry (MS) method for quantifying plasma membrane CFTR. Wild type CFTR, isolated from CFBE41o-, BHK and HT29 cells by immunoprecipitation and cell surface biotinylation, was digested by trypsin and the resulting peptide mixture was analyzed by high performance liquid chromatography (LC)-MS. Several peptides were selected as candidate signature peptides for the protein, based on LC-MS analysis of the tryptic digest of full-length, wild type CFTR. After validation experiments, one peptide (NSILTETLHR, termed as CFTR01) was chosen as the marker peptide for the protein. Compared with the intact CFTR, the marker peptide was a better analyte with regard to sample preparation and MS analysis. An oxygen-18 stable isotope-labeled version of CFTR01 was also prepared and used as the reference peptide, which provided internal standardization for MS quantitation of CFTR in each sample. Quantitation of the marker and reference peptides was performed using multiple reaction monitoring MS. A workflow was established from generating the marker peptide for CFTR samples to MS quantitation. This analytical platform had a low femtomole and high attomole quantitation limit with a statistical error of about 20%. The MS assay of plasma membrane CFTR is expected to be useful in high throughput drug discovery experiments designed to identify compounds that increase the plasma membrane expression of deltaF508-CFTR, as well as in studying cellular trafficking of CFTR.

(Supported by YAO07XX0 to XY and NIH R01-DK/HL-45881 and STANTO7R0 to BAS.) Ubiquitination has multiple cellular functions including targeting proteins for proteasomal and lysosomal degradation. Ubiquitination is a process where the C-terminal glycine residue of ubiquitin covalently attaches to the lysine residue in the target protein through an isopeptide bond. Polyubiquitin chains can be formed by repeated addition of ubiquitin to one of the seven lysine residues in ubiquitin itself. The polyubiquitination chain serves as a signal to tag its substrate protein for proteolysis in the lysosomal or proteasomal compartments, or in some cases to nonproteolytic events. The configuration of the polyubiquitination chain determines in which compartment proteolysis occurs. Lysine (K) 48-linked chains are well established as mediators of proteasomal degradation. Recently, non-classical, lysine (K) 63-linked ubiquitination was found to act in lysosomal degradation and nonproteolytic events. ∆F508-CFTR undergoes proteasome-dependent degradation by passing through ER-associated quality control mechanisms. Wildtype CFTR or rescued ∆F508-CFTR in the cell periphery undergoes lysosome-dependent degradation and is thus also subjected to peripheral quality control. Previously, we have demonstrated that association of wild-type CFTR with CAL leads to the enhanced lysosomal degradation of mature CFTR. To investigate the role of ubiquitination in this process, we epitopically expressed mutant ubiquitins and monitored their effect on CAL-mediated CFTR degradation. To determine whether lysine 48 or lysine 63 linked polyubiquitination is involved in the CAL-mediated degradation of CFTR, we transfected HA tagged wt, K48 or K63 ubiquitin with GFP-CFTR and myc-CAL into HEK293 cells. Both K48 and K63 polyubiquitination of GFP-CFTR were detected by immunoprecipitation of GFP-CFTR followed by Western blotting with an HA antibody. Overexpression of myc-CAL enhanced the degradation of CFTR as expected. Expression of wt and K48 ubiquitin had no effect on CFTR degradation. However, the K63 ubiquitin mutant blocked the degradation of GFP-CFTR caused by myc-CAL. These results suggested that K63-linked polyubiquitination is involved in CALmediated degradation of CFTR in the lysosomal compartment. We are currently utilizing bioinformatics tools and binding assays to identify which E3 ligases bind specifically to the coil-coiled domain of CAL and play a role in the CAL-mediated degradation of CFTR. Supported by the Cystic Fibrosis Foundation and the National Institutes of Health.

Tsai, M. 1, 2 ; Li, M. 2 ; Hwang, T. 1, 2 1. Medical Pharmacology and Physiology, Columbia, MO, USA; 2. Dalton Cardiovascular Research Center, Columbia, MO, USA Cystic fibrosis transmembrane conductance regulator (CFTR) is a phosphorylation-regulated chloride channel gated by ATP binding and hydrolysis at its nucleotide binding domains (NBD1 and NBD2). Mutations in the gene encoding CFTR cause the lethal genetic disease cystic fibrosis (CF). G551D is the third most common CF-related mutation worldwide. We've previously demonstrated that the G551D mutation completely eliminates the ATP-dependent gating of CFTR, rendering the open probability of the mutant channel ~100-fold lower than that of wild-type CFTR (Bompadre et al. J Gen Physiol 2007; 129:285) . The finding that N6-(2-phenylethyl)-ATP (PATP), a high affinity ATP analogue, potentiates G551D-CFTR by binding to NBD1 (Bompadre et al. J Biol Chem 2008; 283:5364) raises the possibility that enhancing nucleotide-NBD1 interaction may alleviate the functional defects. Our studies showing that the lock-open time of CFTR elicited by MgPPi is positively correlated with the ATP bringing stability in NBD1 (Tsai et al. J Gen Physiol 2009; 133:405) allows us to systematically search for candidate mutations that strengthen ATP-NBD1 interaction. Here, we report that a mutation at residue 401 (W401F), which has a lock-open duration ~2-fold longer than WT, turns G551D-CFTR ATP-responsive! In excised membrane patches, the basal current of W401F/G551D-CFTR was increased by 7.78 ± 0.68 (n = 8) fold upon the application of 2.75 mM ATP. PATP (50 µM) increased the basal current by an even greater magnitude (22.34 ± 3.12 fold, n = 5) than in G551D-CFTR (6.14 ± 0.92 fold, n = 8), suggesting that W401F indeed further stabilizes nucleotide binding in NBD1. Interestingly, we found that the enhanced response to nucleotides by the W401F mutation was blunted by the S1347G mutation at NBD2's signature motif as reflected by a decreased ATP/basal ratio of 1.79 ± 0.19 (P < 0.001, n = 12) and PATP/basal ratio of 4.83 ± 0.55 (P < 0.001, n = 5). As crystal structures of NBDs from many ABC proteins show that this serine residue at the signature sequence of NBD2 forms a hydrogen bond with the bound ATP molecule in NBD1, our results suggest that ATP or PATP link together F401 and S1347 from opposite NBDs, bringing the channel into a conformational state that functions better. We propose that reagents being able to promote entering into this state may help improve the functionality of G551D-CFTR. More extensive studies are now underway to characterize the newly-identified functional state and its relationship with pharmacological reagents.

Supported by NIH grants to T.C. Hwang (NIHR01HL53445, DK55835)

BioFocus DPI's proprietary Natural Products library is a unique collection of pre-purified sub-fractions generated from 14,000 individually cultivated actinomycetes and fungi. HPLC fractionation of fermentation broths yielded 140,000 HTS-compatible sub-fractions, each containing on average 10 to 15 compounds. A high-throughput, cell-based assay employing an YFP halide reporter was developed in a patient-derived human bronchial epithelial cell line carrying the ∆F508-CFTR mutation (CFBE41o-), whereby the rate of halide influx is directly correlated to ∆F508-CFTR activity. This assay was bench-marked with the CFFT panel of reference CFTR modulator compounds.

All 140,000 sub-fractions have been screened on the YFP-halide reporter assay. Positive sub-fractions have been subjected to further fractionation and de-convolution to identify single active compounds. The active compounds have been further validated in primary bronchial epithelial cell cultures from CF patients in Ussing chambers.

Screening this untapped resource of novel chemical structures, readily available from the producer strains, and amenable to chemical optimization is believed to lead to the identification of good candidates for the development of new CF therapies. Interestingly, the structures of some of the hits point to several mechanisms of action, such as inhibition of HDACs.

Acknowledgments: We thank Bill Guggino, Ineke Braakman, Kevin Foskett, Stuart McCombie and Eric Gordon for helpful discussions. We acknowledge the support of Cystic Fibrosis Foundation Therapeutics, including their facilitation of the testing of our compounds at ChanTest, Inc. and the guidance of Dr. Melissa Ashlock.

Hoelen, H.; Kleizen, B.; Charitou, P.; Braakman, I. Cellular Protein Chemistry, Utrecht University, Utrecht, Netherlands

The most common mutation in cystic fibrosis (CF) patients, deltaF508 in the first nucleotide binding domain (NBD1), leads to misfolding and degradation of CFTR.

To study its folding we translated CFTR in vitro in the presence of semipermeabilized cells as a source of ER membrane and subjected the protein to mild protease treatment. The resulting protease-resistant fragments represent folded domains of CFTR. Comparison of the proteolytic patterns of wild-type CFTR and the deltaF508 mutant showed the loss of an NBD1related protease resistant fragment in the deltaF508 mutant, which was recovered upon insertion of the I539T suppressor mutation. Limited proteolysis on purified as well as in vitro translated NBD1 domains gave similar results, showing that a 25-kDa fragment within the NBD1 domain lost its stability in the deltaF508 mutant. Using antibodies that recognize specific sequences within the NBD1 domain, we found that the destabilized NBD1 fragment represents the NBD1 core domain, missing both the N-and C-terminus. We concluded that the deltaF508 mutation induced a local conformational change within the NBD1 domain of CFTR that could be reversed with I539T, which effectively rescued NBD1 stability.

This loss as well as rescue of stability occurred already during translation, as soon as NBD1 was synthesized by the ribosome. We now examine how molecular chaperones affect CFTR conformation during the co-translational folding process, and to what extent the rate of synthesis influences folding efficiency and conformation of CFTR. (Wright et al. Physiol Genomics 2004; 16:204) have suggested that 4PBA causes a more global cellular adaptation. We hypothesized that elements of this adaptation would also lead to improved ∆F508-CFTR intracellular trafficking. We performed a differential display RT-PCR screen on RNA isolated from IB3-1 CF epithelial cells treated with 4PBA for 0-24 hours, and isolated a cDNA encoding a putative human STAT-3 (signal transducer and activator of transcription-3) interacting protein (StIP-1) that regulates STAT-3. StIP-1 is identical to Elongator protein 2 (Elp2); Elongator is a protein complex that facilitates mRNA transcription by RNA polymerase II, and may also have a role in regulation of polarized vesicular transport (Rahl et al. Mol Cell 2005; 17:841) .

4PBA causes transiently increased StIP-1 mRNA and protein abundance in IB3-1 CF epithelial cells after 8 hours of exposure. In immunofluorescence experiments, overexpression of StIP-1/Elp2 in IB3-1 cells increased deltaF508-CFTR trafficking to the plasma membrane even in the absence of 4PBA treatment. In contrast, overexpression of a mutant of StIP-1/Elp2 blocked the 4PBA-stimulated improvement of deltaF508-CFTR trafficking. StIP-1/Elp2 colocalizes with a 58k Golgi protein (Golgi marker) as well as with Snap25/23 (a marker of trafficking vesicles). These data suggest that StIP-1/Elp2 may regulate deltaF508-CFTR trafficking along the exocytic pathway.

As Elongator also directly regulates Hsp70 expression in yeast (Han et al. Acta Biochim Biophy Sin (Shanghai) 2007; 39 :453), we tested whether 4PBA may regulate Hsp70 expression via StIP-1/Elp2. In T84 cells transfected with siRNA targeting StIP-1/Elp2, StIP-1/Elp2 protein expression was reduced by about 50% and Hsp70 expression was reduced by 80% compared with control cells. These data suggest that StIp-1/Elp2 may also influence ∆F508-CFTR trafficking by regulating Hsp70 expression.

∆F508 CFTR is recognized as a misfolded protein by the ER quality control mechanism and targeted for degradation by the proteasome. Transcomplementation of ∆F508 CFTR by truncated forms of CFTR have been observed to overcome defective processing and allow mature ∆F508 CFTR to reach the cell surface. Here we study the combined roles of a truncated form of CFTR; ∆264 CFTR which is missing the first 4 transmembrane segments of CFTR and small molecule correctors or proteasome inhibitors in the rescue of ∆F508 CFTR.

We observed in transiently transfected Cos7 cells that ∆264 CFTR protein expression was significantly increased by proteasome inhibitors, correctors and low temperature. This suggests that ∆264 CFTR and ∆F508 CFTR are processed by similar pathways in the cell.

Previously, we showed that ∆264 CFTR enhanced the maturation of ∆F508 from B to C bands by transcomplementation.

In this study, we observed that the correctors 4A and VRT325 increased the maturation of ∆F508 CFTR from B to C bands. The goal of this study is to see if there is a larger effect on ∆F508 CFTR by combining the truncated form with correctors or proteasome inhibitors.

In cells cotransfected with ∆F508 CFTR and ∆264 CFTR and treated with increasing doses of either 4A or VRT 325, there was a significant increase in the mature C band of ∆F508 CFTR compared to the cells cotransfected with ∆F508 and ∆264 CFTR in the absence of correctors or cells transfected with ∆F508 CFTR alone.

Previously we showed that the proteasome inhibitors alone could increase the maturation of B to C bands of ∆F508 CFTR. In this study we tested the effect of combining transcomplementation with proteasome inhibitors. In cells cotransfected with ∆F508 CFTR and ∆264 CFTR and treated with increasing doses of either MG132 or PS 341, there was a larger increase in the mature C band and a dramatic decrease of B band of ∆F508 CFTR compared to the cells cotransfected with ∆F508 and ∆264 CFTR in the absence of proteasome inhibitors or cells transfected with ∆F508 CFTR alone.

In conclusion: Rescue of ∆F508 CFTR occurs when cells are treated with the small molecule correctors 4A, VRT 325 or when cotransfected with ∆264 CFTR. However, a combination of gene therapy with this truncated form of CFTR and small molecule correctors or proteasome inhibitors is more effective in rescuing the maturation of the C band of ∆F508 CFTR. The greatest effect is seen when proteasome inhibitors are combined with ∆264 CFTR than when the small molecule correctors are combined with ∆264 CFTR.

Muimo, R. Academic Unit of Respiratory Medicine, University of Sheffield, Sheffield, United Kingdom bilized wt-CFTR at the plasma membrane. As detected by cell surface biotinylation followed by CFTR immunoprecipitation, co-expression of CK19 increased the plasma membrane expression of wt-CFTR 2.5-fold. In Calu-3 and HeLa cells stably expressing wt-CFTR, the over-expression of CK19 using recombinant adenovirus (AdCK19) produced ~2-fold increase in CFTR-mediated Clfluxes/currents, assayed by SPQ and whole-cell patch clamp. The physiological role of CK19 was examined using recombinant lentiviruses containing shRNAs targeting its expression in HEK293 cells. Knockdown of endogenous CK19 by 50% yielded an 85% reduction in wt-CFTR at the plasma membrane. Cell surface CFTR internalization assays indicated that the expression of CK19 reduced the rate of wt-CFTR endocytosis. Thus, inhibition of CFTR endocytosis is a key aspect of the mechanism by which CK19 promotes apical CFTR stability. Pulse-chase experiments indicated that CK19 expression enhanced the maturation efficiency of wt-CFTR, while having no significant effect in promoting the maturation of ∆F508-CFTR. Interestingly, however, the over-expression of CK19 stabilized temperature rescued ∆F508-CFTR at the plasma membrane, increasing its density by 4.6-fold. This result was supported by transepithelial current measurements across CFBE41o-∆F508 epithelia, which showed that CK19 expression potentiated forskolin-stimulated Clcurrent by 65% after temperature rescue of the mutant. Taken together, our data reveal a novel mechanism by which the intermediate filament protein, CK19, stabilizes CFTR at the plasma membrane, increasing its plasma membrane density by inhibiting CFTR endocytosis. The retention of rescued ∆F508-CFTR at the plasma membrane by CK19 may provide a therapeutic approach to CF by stabilizing mutant protein that escapes ER-associated degradation, or by facilitating the actions of ER-targeted correctors of ∆F508-CFTR trafficking.

[Supported by the CFF SUN08G0, and NIH grants DK68196 and DK72506.] The most common mutation in cystic fibrosis (CF), F508del, results in CFTR (CF transmembrane conductance regulator) protein that is retained in the endoplasmic reticulum by a quality control system (ERQC). Previously, we have shown that miglustat (N-butyldeoxynojirimycin, Zavesca®), an inhibitor of the alpha-1,2-glucosidase, corrects the defective trafficking of F508del-CFTR and hypothesized that by inhibiting the interaction of F508del-CFTR with calnexin, miglustat prevents its retention and degradation [1] . However, others contest the role of calnexin in the F508del-CFTR retention [2] . The purpose of this study was i) to determine the effect of calnexin small interfering RNA (siRNA) on the trafficking of endogenous F508del-CFTR; ii) to compare these results with a miglustat-induced correction; and iii) to understand whether calnexin is implicated in the F508del-CFTR trafficking.

Human tracheal gland epithelial cells (CF-KM4) were transfected by siRNA calnexin (0.5µg/ml, 72h), siRNA control (0.5µg/ml) or were treated with miglustat (100µM, 2h). The level of calnexin expression was verified by biochemistry and consequences on CFTR and ENaC channel activities were assessed using single-cell fluorescence imaging. We compared the results with those obtained on untreated-and reverted-(CF-KM4 stably transfected with the CFTR wild type) CF-KM4 cells. We demonstrated that transient transfection of calnexin siRNA significantly knocked down calnexin expression (~ 75%) and resulted in a rescue of F508del-CFTR activity at the plasma membrane and a decrease by 66% of ENaC activity. Miglustat induced a 1.5 times upper correction of F508del-CFTR and ENaC function in CF-KM4 corresponding to the level of activities measured in reverted CF-KM4. Moreover, similar level of CFTR and ENaC activities was also recorded in CF-KM4 cells simultaneously treated with miglustat and transfected by calnexin siRNA.

In conclusion, this work is in favor of a role of calnexin in miglustatinduced correction and confirms calnexin as a valuable CF therapeutic target. However, our results suggest that inhibition of calnexin/F508del-CFTR interaction is probably not sufficient to fully explain the effects of miglustat, raising the hypothesis that another molecular target for this drug exists.

References:

1. Norez C et al. Rescue of functional delF508-CFTR channels in cystic fibrosis epithelial cells by the alpha-glucosidase inhibitor miglustat. FEBS Lett 2006; 580:2081 -2086 2. Okiyoneda T et al. Role of calnexin in the ER quality control and productive folding of CFTR; differential effect of calnexin knockout on wildtype and DeltaF508-CFTR. Biochim Biophys Acta 2008; 1783 :1585 The interactions between the domains of the cystic fibrosis transmembrane conductance regulator (CFTR) are important for its function as a chloride channel and many mutations leading to cystic fibrosis (CF) are thought to disrupt critical intramolecular interactions. NBD2 is a critical component of these interactions, including the NBD1/NBD2 dimer required for ATP hydrolysis and channel opening. Multiple models of CFTR [1] [2] [3] point to significant interactions with the intracellular loop (ICL) regions that are likely important for transmitting information about the NBD state to the channel pore. Additionally, intriguing hints in the literature suggest that the R region, whose phosphorylation state regulates dimer formation and channel function [4] , also interacts with NBD2 [5, 6] . It has been difficult to obtain isolated NBD2 reagent for biochemical and structural studies, although a crystal structure has been determined for a NBD2-MalK fusion protein. Even with five solubilizing mutations, the isolated human NBD2 (hNBD2; residues 1193-1445) is only marginally soluble. However, by optimizing the purification protocol and buffer conditions, we have obtained sufficient hNBD2 concentrations for study via NMR spectroscopy.

Based on this significant achievement, we are currently performing NMR paramagnetic relaxation enhancement (PRE) experiments to probe for the presence of the weak, transient NBD1/NBD2 binding. Additionally, we plan to test NBD2 interactions with ICL peptides and with the R region in both unphosphorylated and phosphorylated states using NMR nitrogenproton correlation experiments. The results should provide insights into NBD2 interactions with NBD1 and other parts of CFTR that are critical for regulated channel function.

stabilities [3] . Consequently, we present subsequent studies that were conducted to reconcile these different observations. Thermal stabilities were studied by differential scanning calorimetry (DSC), which measures the heat absorbed during the protein unfolding process. Thermal unfolding of wild type (residues 387-646(∆405-436)) and ∆F508 NBD1 was performed as a function of urea concentrations of 0 -6 M. Structural changes occurring through the thermal transition were monitored by near and far UV circular dichroism. Using a non-linear Levenberg-Marquardt algorithm to globally fit the data, the DSC transitions can be fit to a multi-step unfolding pathway that includes an irreversible step. The accompanying poster of Hunt and coworkers provides complementary isothermal unfolding data under conditions which overlap with those studied by DSC. Together, these studies enable us to propose a global unfolding model for NBD1.

1. Qu BH, Thomas PJ. Alteration of the cystic fibrosis transmembrane conductance regulator folding pathway. J Biol Chem 1996;271:7261-7264.

2. Lewis HA et al. Impact of the deltaF508 mutation in first nucleotidebinding domain of human cystic fibrosis transmembrane conductance regulator on domain folding and structure. J Biol Chemistry 2005;280:1346-1353.

3. Brouillette C et al. Differential scanning calorimetry detects a difference in the stability of the ∆F508 mutant of the nucleotide binding domain 1 of CFTR. Pediatr Pulmonol 2008;S31:227.

Financial support from the CFF and technical assistance from Ms. Kris Connors is gratefully acknowledged.

In epithelial cells CFTR is endocytosed rapidly from the apical plasma membrane. The majority of internalized CFTR is then recycled efficiently back to the cell surface through the endosomal recycling compartment. The aim of the present study was to investigate the role of the recently cloned lemur-tyrosine kinase (LMTK2) in regulating the recycling of CFTR in polarized human epithelial cells. RT-PCR analysis of human lung, intestine and pancreatic tissues revealed the presence of LMTK2 transcript in all tissues. Image analysis revealed that LMTK2 localizes to early and recycling endosomes and shows overlap with CFTR in these compartments. siRNA mediated knock-down of LMTK2 in T84 cells caused CFTR to be trapped in endosomes, reducing apical membrane CFTR levels and dramatically reducing forskolin-stimulated chloride secretion. In contrast, over expression of LMTK2 led to enhanced CFTR recycling and an increase in forskolin-stimulated chloride movement. Since LMTK2 is a kinase, we generated a point-mutation in the active site, rendering the kinase catalytically inactive. Overexpression of the "dead kinase" resulted in a loss of CFTRmediated chloride currents, similar to that seen with knock-down of LMTK2, suggesting that the dead kinase acts as a dominant negative. Furthermore, we show that CFTR is a target substrate for LMTK2 kinase activity. These studies demonstrate a role for LMTK2 in the recycling of CFTR in polarized epithelia, and suggest that phosphorylation events are involved in the acute regulation of CFTR trafficking. Finally our results suggest that activation of LMTK2 could be therapeutically beneficial by enhancing the recycling of internalized ∆F508 CFTR.

Drevillon, L. 1 ; Tanguy, G. 1 ; Hinzpeter, A. 1 ; Arous, N. 1 ; de Becdelievre, A. 1, 2 ; Fanen, P. 1, 2 1. U955, INSERM, Creteil, France; 2. Faculté de Médecine, Université Paris 12, Creteil, France Cystic fibrosis (CF) is mainly caused by mutations interfering with the biosynthetic folding of the CFTR (cystic fibrosis transmembrane conductance regulator) protein. To better understand the regulation of CFTR processing and trafficking, we made a genetic screen, which identified COMMD1 as a new CFTR partner. COMMD1 was first found to be involved in the regulation of hepatic copper excretion and then in sodium uptake through interaction with ENaC (epithelial sodium channel). Currently, a growing number of data suggests that COMMD1 is associated with the ubiquitin proteasome system for regulating protein stability of NF-κB subunits, ATP7B and HIF-1α.

We provide herein the first evidence that i) COMMD1 interacts with CFTR in cells expressing endogenously both proteins; ii) COMMD1 promotes CFTR cell surface expression as assessed by biotinylation experiments in heterologously expressing cells; and iii) COMMD1 regulates CFTR ubiquitinylation. Altogether, our data and the review of the literature suggest a mechanism by which COMMD1 can act as a scaffold protein contributing to protein ubiquitinylation in the endocytic/recycling pathway. The fate of the proteins interacting with COMMD1 varies, leading either to degradation or to modified trafficking. We suggest a general function in controlling ubiquitinylation of proteins implicated in various cellular processes.

A main feature of CF is an abnormal balance between fluid and electrolyte transport, leading to reduced chloride secretion combined with an abnormal sodium absorption in pulmonary epithelium. COMMD1 inhibits ENaC absorption and enhances CFTR trafficking. Together, these data may open new therapeutic avenues to improve CF patients' care.

This work was supported by public grants from INSERM, Chancellerie des Universités de Paris and the French Association "Vaincre La Mucoviscidose." Misfolding and premature degradation of the mutant Clchannel CFTR∆F508 causes cystic fibrosis (CF). CFTR∆F508 biogenic defects are conditional and folding correctors are actively being developed as CF therapeutics. How the cellular environment impacts CFTR∆F508 folding efficiency, as well as the identity of correctable folding defects in CFTR∆F508, is currently unclear. The folding progression of CFTR appears to be monitored by at least two distinct endoplasmic reticulum quality control (ERQC) complexes that are located in the ER membrane and cytosol. The membrane-associated RMA1/Derlin-1 complex appears to recognize folding defects in CFTR that may involve misassembly of MSD2 into a complex with the amino-terminal regions of CFTR (1) . In contrast, the cytosolic Hsc70/CHIP complex appears to sense folding defects that are related to misassembly of NBD2 (1) . However, these previous observations were made with over-expressed RMA1 or CHIP and the extent to which endogenous levels of the ERQC components determine the fate of nascent forms of CFTR∆F508 has not been well-defined. In addition, whether activity of the ERQC machinery impacts the effectiveness of drugs developed to correct misfolding of CFTR∆F508 is unknown. We demonstrate that decreasing the endogenous pools of the E3 ubiquitin ligases RMA1 or CHIP by siRNA permits a sub-population of CFTR∆F508 to fold, escape the ER and accumulate as a maturely glycosylated species. In the absence of these ERQC factors, we found folding corrector Corr-4a to be more effective at rescuing folding defects in CFTR∆F508. Amazingly, the combination of RMA1 inactivation and Corr-4a treatment dramatically enhanced CFTR∆F508 folding 13-fold while a 6-fold increase was observed with Corr-4a treatment alone. It appears that some, but not all, folding defects in CFTR∆F508 are correctable, and a large pool of nascent CFTR∆F508 is degraded before Corr-4a action. RMA1 recognizes defects in CFTR∆F508 related to misassembly of a complex that contains MSD1, NBD1 and the R-domain (2) . Corr-4a could not rescue ∆F508 dependent folding defects in these amino-terminal regions, but instead acts on CFTR∆F508 after MSD2 synthesis. Corr-4a appears to act in part via stabilization of the MSD2, which may enhance critical contacts between MSD2 and other sub-domains of CFTR, such as MSD1 or NBD1. Correction of folding defects recognized by the RMA1 ubiquitin ligase and/or global modulation of ERQC activity has the potential to increase CFTR∆F508 folding and provide an approach to treat CF.

Supported by the Cystic Fibrosis Foundation.

whether functional activation of the dF508 CFTR could be efficiently assessed in primary CF AECs. Methods: Nasal epithelial cells and AECs were obtained from dF508 homozygous CF patients (mean age: 3.11±2.25 y.o) by non-bronchoscopic brushing. dF508 CFTR correction was examined by adopting the yellow fluorescent protein (YFP) halide reporter assay (Galietta et al. Am J Physiol Cell Physiol 2001; 281:1734) . Cells were transduced in 96-well plates with YFP adenovirus at specific titer (MOI) and optionally with wild type (WT) or dF508 CFTR adenoviral vectors. For corrector controls, cells were transduced with shRNA either against STX-8 or BCAP31, in order to knockdown mRNA and/or protein expression of syntaxin-8 (STX-8) or B-cell receptor associated protein 31 (BCAP31), followed by YFP transduction. These two molecules are involved in the trafficking of CFTR (Fischer et al. Ped Pulmonol 2006; 41(S29):209) . Cells were then pre-incubated with 100µM genestein and 10µM forskolin, wells injected with iodide and YFP signal examined over 12 seconds. Correction of dF508 CFTR was determined by the quenching of YFP signal.

Results: Our results demonstrated that CF AECs were successfully transduced with adenovirus at >75% efficiency with no observed morphological cytotoxic effects. Knockdown of STX-8 and BCAP31 expression enhances correction of dF508 CFTR in CF AECs. A gold standard method of temperature rescue whereby the cells were incubated at 27∞C was conducted using CF nasal epithelial cells and the results confirmed restoration of dF508 CFTR function in these cells. Having validated this technique on primary cultures, we have performed halide assays assessing the pharmacotherapeutic potential of several class compounds identified from preliminary work utilizing bronchial epithelial cell lines. Data produced has identified a number of agents capable of restoring CFTR function in primary CF AECs.

Conclusions: These findings provide proof of principle that correction of CFTR function in CF AECs can be efficiently assessed using YFP-halide reporter assay. Furthermore, this is the first data to demonstrate CFTR restoration in primary CF lower AECs using a small molecule discovery approach. In light of this, restoration of dF508 CFTR function can be assessed by use of pharmacological agents or other molecules/proteins that might play a role in CFTR or CFTR-dependent signaling pathway(s). Therefore, this novel technique will provide insights into new therapies for the treatment of cystic fibrosis.

Karamyshev, A.L. 1 ; Patrick, A.E. 1 ; Johnson, A.E. 2 ; Thomas, P.J. 1 1. Physiology, UT Southwestern Medical Center at Dallas, Dallas, TX, USA; 2. Texas A&M Health Science Center, College Station, TX, USA Cystic fibrosis (CF), a common human genetic disease, is caused by mutations in CFTR (Cystic Fibrosis Transmembrane Conductance Regulator). Over 1,600 mutations throughout the CFTR gene have been identified, many of which disrupt the function of the CFTR protein to cause CF. CFTR is a member of the ABC transporter superfamily that transports chloride ions across the apical membrane of epithelial cells. During biogenesis, the synthesis of CFTR polypeptides initiates on cytosolic ribosomes that are then targeted to the translocon in the ER membrane where the nascent CFTR chain is integrated into the membrane prior to transport through the Golgi to the plasma membrane. Interactions with many other proteins are essential to these processes. Some CF-associated mutations likely alter cotranslational interactions of the CFTR nascent chain with the protein factors involved in its folding, transport, modification, quality control, etc. In the current work, cotranslational protein interactions with the CFTR nascent chain have been identified using site-specific photocrosslinking. This method is based on tRNA-mediated incorporation of a single photo-reactive probe into the CFTR nascent chain at specific positions during its synthesis in vitro. Amber suppressor N ε -(5-azido-2-nitrobenzoyl)-Lys-tRNA amb (εANB-Lys-tRNA amb ) is used to incorporate the probe into the nascent chain wherever an amber stop codon is introduced into the CFTR mRNA. Ribosome-bound CFTR nascent chains of a defined length were prepared by translation of truncated CFTR mRNAs. During the process of elongation, the extreme N-terminus of the CFTR nascent chain is found adjacent to 65, 80, 90, [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110] 50 , and 70 kDa proteins depending on the length of the nascent chain and the location of the probe. Analysis of the photocrosslinking patterns demonstrates that the CFTR nascent chain is transferred from one protein to another during the process of its translation. To identify these proteins, a system was developed that includes: 1) in vitro preparation of CFTR ribosome nascent chain complexes with the photocrosslinking probe; 2) fractionation of cellular proteins; 3) selection of fractions containing the protein of interest using a photocrosslinking screen; and 4) identification of proteins in the selected fractions by mass spectrometry or immunoprecipitation. The first set of three interacting proteins identified includes: an N-terminal acetyltransferase-like protein (NAT), a Signal Recognition Particle subunit (SRP54), and the translation elongation factor, eEF-1α. Notably, wild-type and mutant CFTR exhibit different interaction efficiencies with some of these proteins, which may provide a mechanism of sorting wild-type from mutant CFTR.

(Supported by NIH NIDDK and the CFF.)

The predominant mutation causing cystic fibrosis (CF) in the human population is the deletion of phenylalanine residue 508 (deltaF508) in the first nucleotide-binding domain (NBD1) of the cystic fibrosis transmembrane conductance regulator (CFTR). While the deltaF508 mutation alters the surface topography of NBD1 at its likely site of attachment to the transmembrane domains of CFTR, it does not alter the overall fold of the domain. However, there has been controversy concerning the influence of the mutation on the folding properties and stability of NBD1. Different biophysical techniques have given inconsistent results regarding whether the mutation significantly destabilizes the domain. This controversy has created uncertainty concerning the molecular defects that cause the pathological retention of deltaF508 CFTR molecules in the endoplasmic reticulum. To clarify the influence of the deltaF508 mutation on the thermodynamics of NBD1 folding, we have been characterizing its unfolding pathway using a combination of isothermal (this abstract) and thermal (accompanying abstract by Brouillette and coworkers) denaturation studies. The latest status of these studies and their implications for the molecular etiology of CF will be discussed. This work was supported by a grant from the CFFT. In Sept. 2007 AMRI entered into collaboration with Cystic Fibrosis Foundation Therapeutics to screen AMRI's collection of natural product samples, for correctors and potentiators of the ∆F508-CFTR response. AMRI's Natural Products Libraries comprise nearly 300,000 fermentations, extracts, and fractionations from over 35,000 unique plant, fungal, actinomycete, microorganism, and marine invertebrate sources. Natural products have been a rich source for drug leads. Between 1981 and 2002, 52% of all NCEs (new chemical entities) originated as natural products or were synthetic derivatives based on natural products. High-throughput screening (HTS) of the AMRI libraries was performed using a YFP/∆F508 CFTR FRT cell line in a 384-well FLIPR TetraPlus TM NaI Flux assay. The initial HTS yielded 8620 corrector and 4171 potentiator hits which exhibited activity at least 3 standard deviations above the mean of the blank control. Hit confirmation was performed by four-point dose/response in the NaI Flux screening assay. Samples with non-specific activity due to intrinsic sample fluorescence or which caused disruption of cellular membrane integrity were eliminated by performing four point dose/responses in the NaI Flux assay in the presence of CFTRinh-172, a CFTR blocker. Confirmed hits were prioritized based on their activity, dose/response effect, and lack of toxicity. Samples from 92 unique sources were further selected for dereplication and structure identification. Purified compounds were tested for corrector and potentiator activity in the NaI Flux assays and in 24-well corrector and potentiator conductance assays developed by Dr. Robert Bridges. One of the active compounds identified from a bacterial source was genistein, a well known potentiator of ∆F508 CFTR. Dereplication of additional fractions from other sources has yielded unique active scaffolds as possible leads for drug development of CFTR correctors and potentiators.

AMRI gratefully acknowledges the support from CFFT for this project. We also thank Dr. Robert Bridges for his assistance with the conductance assays.

Sokolow, A.G. 1 ; Irvin, J. 2 ; Corey, D.A. 2 ; Kelley, T.J. 2 1. Pediatric Pulmonary, Rainbow Babies and Children's Hospital, Cleveland, OH, USA; 2. Pediatrics, Case Western Reserve University, Cleveland, OH, USA Previously we identified that aspects of the cAMP pathway indicate a chronic activation of the inflammatory signaling cascade in cystic fibrosis (CF) epithelial cells. This increase in cAMP signaling was found to be responsible in part for the perinuclear accumulation of cholesterol as intervention with the cAMP-binding competitor Rp-cAMPS reverts cholesterol transport to wt characteristics. The hypothesis of this study is that a chronic increase in cAMP signaling is an initiating factor in CF-related inflammatory signaling. Markers of inflammatory signaling we have identified as exhibiting CF-related changes are an increase in STAT1, a decrease in SMAD3, and a decrease in NOS2 protein expression. CF-model 9/HTEocells were treated with Rp-cAMPS (50 µM) for 72 h and examined for STAT1 and SMAD3 protein expression by Western blot. STAT1 protein expression decreases significantly, while SMAD3 expression significantly increases in CF-model cells in the presence of Rp-cAMPS. These data demonstrate a reversion to wt phenotypes in the presence of Rp-cAMPS. To assess the effects of Rp-cAMPS in vivo, ∆F508 mice (∆F/∆F) were treated with either saline or Rp-cAMPS (5mg/kg/day) for 12 days. Nasal epithelium was excised and examined for NOS2 and SMAD3 protein expression. Mice treated with Rp-cAMPS show restored NOS2 and SMAD3 expression in CF mice, while control mice demonstrate limited expression. These data show that the cAMP feedback response is critical to the control of CF-related signaling and that it is reversible with Rp-cAMPS. In order to assess whether the cAMP pathway has a potential impact on CF inflammatory signaling, the effect of Rp-cAMPS on NF-κB activation and IL-6 and IL-8 production in response to IL-1β and TNF-α was examined. NF-κB activation as measured by luciferase (NF-κB-luc) is diminished by Rp-cAMPS in CFmodel 9/HTEo-cells but not affected in control cells. These data suggest a CF-related bias in the regulation of NF-κB signaling by the cAMP pathway. If there is cAMP-mediated regulation of NF-κB activation in CF cells, then it is expected that cytokine production should be similarly affected by Rp-cAMPS as was NF-κB. To test this prediction, IL-8 and IL-6 secretion was examined in CF-model 9/HTEo-cells and controls in response to IL-1β and TNF-α. Identical to NF-κB activation results, Rp-cAMPS has little influence on IL-8 or IL-6 production in control cells, but significantly reduces the production/secretion of these cytokines in CF-model cells. These data further demonstrate the central role of cAMP-related signaling as a key step in a number of CF cell regulatory characteristics. In vivo examination of this anti-inflammatory effect of Rp-cAMPS in CF mouse models is underway. Mechanistic analysis suggests that Rp-cAMPS-mediated control of CREB binding protein (CBP) binding to the p65 subunit of NF-κB is responsible for the anti-inflammatory properties of Rp-cAMPS in CF models. We conclude that chronic activation of the cAMP pathway in CF alters cell regulatory balance to favor pro-inflammatory signaling and that this balance is reverted to wt characteristics with Rp-cAMPS. This work is supported by the NIH/NHLBI and CFF. Sci USA 2008; 105:4335-9) . We further showed that these cells are reconditioned through various signaling pathways (Makam et al. Proc Natl Acad Sci USA 2009; 106:2779-83.) , leading to novel competences. However, these results did not address whether reconditioned airway neutrophils retain immune competences expected from professional phagocytes, e.g., functional innate pathways that have evolved for pathogen elimination. Indeed, neutrophilic inflammation in CF airways coexists with chronic bacterial / fungal infections, suggesting immune incompetency.

Results: Here, we used novel, high-content flow and image cytometry tools to further investigate the immune competence of CF airway neutrophils, as compared to their blood counterparts. First, we assessed their ability to activate the caspase-1 dependent inflammasome, using a specific fluorescent substrate for the enzyme. Upon direct analysis (no exogenous activation) by flow cytometry (4-laser LSRII digital FACS, BD Biosciences), we observed a marked upregulation of caspase-1 activity in CF airway neutrophils, suggesting adequate inflammasome function in these cells. Second, we assessed the activation within CF neutrophils of the hypoxia-inducible factor 1 alpha pathway, a transcription factor pivotal to phagocyte function. Using direct single-cell analysis with a novel image cytometry platform (ImageStream, Amnis Corp.), we observed marked translocation of this transcription factor into the nucleus of CF airway, but not blood, neutrophils. Third, we assessed the ability of CF airway neutrophils to phagocytose bacteria and undergo proper phagosome acidification, a process suggested by some to be altered in CF. Using a rapid activation protocol (60 minutes, at 37 ᭺ C) combined with bacterial particles bearing a pH-dependent fluoroscent label, we demonstrate by flow and image cytometry that CF airway neutrophils are competent at phagosome formation and acidification. Fourth, we sorted live airway neutrophils from the complex mixture of cells and debris present in CF sputum using a newly developed protocol and assessed their mRNA expression profile by microarrays. The mRNA expression profile results obtained are consistent with the notion that, while being reconditioned, CF airway neutrophils do retain essential innate immune functions.

Conclusions: Overall, our functional, multi-platform data suggest that the coexistence of neutrophils and bacteria / fungi in CF patients with established disease may be due more to the selection of immune-resistant pathogen strains than to significant immune incompetency on the part of host neutrophils. However, this does not exclude important deleterious effects linked to some anomalous functions of CF airway neutrophils, e.g., their active release of elastase. Support: CF Foundation, Stanford CHRP Program and Skippy Frank Foundation.

response to TLR ligands for TLRs 2, 3, 4, and 5 . Next, we showed that IL-1β and TLRs increased IL-8 production in CFBE41o-and CFBE41o-/WT cells. The addition of a selective inhibitor of EGFR phosphorylation (AG 1478) decreased IL-1β-and TLR-induced IL-8 production, implicating EGFR activation. Future experiments will explore the communication between TLR and EGFR via a signaling cascade that we have studied in normal airway epithelial cells that includes: NADPH oxidase, production of reactive oxygen species, and metalloprotease cleavage of EGFR ligand.

Conclusions: Here we show that CF airway epithelial cells (CFBE41o-) have increased EGFR-p and IL-8 production compared to CFTR corrected cells (CFBE41o-/WT). In addition, these experiments implicate EGFR signaling in IL-8 production in CFBE41o-cells, suggesting the importance of exploring EGFR signaling in CF as a potential novel therapeutic target.

Supported by NIH and CVRI Research Funds. The zinc finger protein A20 is regulated by and responsible for the termination of NF-κB signalling through immune receptors including Toll-like receptors (TLRs). Mice deficient in A20 display persistent activation of NF-κB by TLRs and tumour necrosis factor receptor, causing mortality among neonates and multiorgan inflammation in adults [1] . Prolonged Pseudomonas aeruginosa (PA)-induced signaling via TLR2 contributes to chronic inflammation in cystic fibrosis (CF) . A20 has been shown to be a negative regulator of TLR2-induced inflammatory responses in the airway [2] . We propose that in the CF epithelium, defective elimination of the TLR2-PA signalling complex, coupled with A20 inhibition leads to enhanced pro-inflammatory signalling through the NF-κB pathway.

Methods: Non-CF (HBE) and CF (CFBE) bronchial epithelial cell lines were grown to 80% confluency on collagen-coated plates prior to stimulation with 50µg/ml PA-lipopolysaccharide (LPS; Sigma). A20, TLR2 and NF-κB (P50 and P65) levels were measured by flow cytometry, 0-12h after stimulation. Nuclear extracts for NF-κB staining were obtained using a DNA CycleTest kit (BD Biosciences). IL-8 release was measured by ELISA (Peprotech) using supernatants from cells treated with PA-LPS for 24h. Statistics were calculated by ANOVA, with n=3-6 for all experiments.

Results: IL-8 release was significantly higher in CFBE cells than HBE cells following 24h stimulation with PA-LPS (P<0.05). Consistently, intracellular expression of TLR2 was higher in CFBE cells than in HBE cells (8h and 12h after PA-LPS exposure, P<0.05). Despite increased intracellular A20 expression 4h (P<0.001) after PA-LPS exposure, expression of A20 in CFBE cells fell below basal levels by 12h exposure (58% of basal expression), while HBE cells showed sustained A20 expression. Basal (2 and 4h after stimulation, P<0.001) . P65 expression increased significantly in both HBE and CFBE cells with exposure to PA-LPS (P<0.05-0.001). However, these increases were more pronounced in CFBE cells.

Conclusions: In chronically inflamed CF epithelium, A20 signalling is significantly inhibited causing increased NF-κB activation and subsequent pro-inflammatory signalling. This work was supported by a grant from the CF Trust UK (PJ541). References: 1. Lee et al. Science 2000; 289:2350 -2354 . 2. Gon et al. Am J Respir Cell Mol Biol 2004 31:330-336 .

Bruscia, E.M. 1 ; Satoh, A. 2 ; Zhang, P. 3 ; Caputo, C. 1 We have shown that CFTR-/-(CF) bone marrow (BM)-derived murine macrophages contribute to the elevated production of various pro-inflammatory cytokines in response to LPS (Bruscia et al. Am J Respir Cell Mol Biol 2009; 40:295-304) . LPS signals through the Toll-like receptor 4 (TLR4) through multiple mechanisms. First, TLR4 signals via the plasma membrane by promoting the activation of NFkB and MAPK (ERK1/2, p38, JNK) cascades. Additionally, the internalization of the TLR4 receptor and its localization in early endosomes activates the interferon beta (IFN-b) response through IRF-3. Subsequently, internalization of TLR4 has been associated with its own degradation and resolution of the inflammatory response. Tight control of active TLR4 is critical for an adequate and controlled inflammatory response. Given that the inflammatory response in CF appears to be hyperresponsive, we investigated the signal transduction and trafficking of TLR4 in WT and CF primary murine macrophages during LPS stimulation.

Macrophages were differentiated from BM of age/sex matched WT (n=3) and CF (n=3) mice. Cells were treated for 10, 20 and 60 minutes with LPS and the protein lysates were assessed for the content of key phosphorylated proteins involved in signal transduction of the three pathways (pIKBa, pERK1/2, p38, JNK and pIRF-3). The phospho-protein content was analyzed by either Western blot or Luminex phospho-protein analysis. Protein lysates from untreated WT and CF cells served as controls. The data suggest that CF macrophages have a more rapid and prolonged signal transduction through IKBa, ERK1/2 and IRF3 compared to WT cells. To study the dynamics of plasma membrane TLR4 internalization and its localization with endosomal compartments, CF and WT macrophages were labeled at the membrane with a fluorescent antibody anti-TLR4 and internalization was induced by incubation at 37 ᭺ C for 5, 15, 30 and 60 minutes in the absence or presence of LPS. We followed TLR4 internalization by flow cytometry, and TLR4 localization with the endosomal compartment by immunofluorescence microscopy (IF) and Imagestream technology. We found that TLR4 internalization was similar in WT and CF macrophages in the absence of LPS such that 50% of the receptors are internalized over 15 min. In contrast, in the presence of LPS, TLR4 internalization was significantly more robust (50% of TLR4 is internalized in less than 5 min) in CF than WT macrophages (p=0.03). In addition, the IF and Imagestream data show that at 30 minutes, CF macrophages had significant accumulation of internalized TLR4 in the endosomal compartment compared to WT cells. In conclusion, our data support the hypothesis that the lack of functional CFTR in macrophages affects the internalization and trafficking of TLR4 during LPS activation and these defects lead to abnormal increase of signal transduction with inefficient resolution. Center, University of North Carolina, Chapel Hill, Chapel Hill, NC, USA In cystic fibrosis (CF), the depletion of upper airway surface liquid (ASL) and the failure of mucus transport leads to the characteristic phenotype. However, the ultimate origin of the ASL is not completely known. One hypothesis is that the alveoli, presumably alveolar type II cells, possess the ability to secrete fluid, and a portion of this liquid from the alveolar space is transported out into the airways, thus, contributing to total ASL volume. Moreover, recent studies have shown that within the upper airway, purine nucleotides, such as ATP/UTP, significantly regulate ion transport properties and airway surface liquid height across the epithelium. However, the role these soluble mediators play to mediate ion transport and liquid homeostasis within the alveoli is still unclear.

To identify the capacity of ion transport within the alveoli, normal and CF human alveolar type II cells were isolated, cultured, and characterized for purity. Bioelectric analysis showed primary normal and CF cultures displaying similar transepithelial resistance, however, CF basal short circuit current was markedly less compared to normal alveolar type II cultures. Treatment of normal cultures with forskolin, the cAMP-inducer, UTP, or ionomycin, significantly induced Clsecretion in the presence of amiloride. Interestingly, forskolin and ionomycin-mediated Clsecretion was significantly reduced in CF alveolar type II cultures in the absence or presence of amiloride. Moreover, normal alveolar type II cultures treated with the extracellular nucleotide, UTP, exhibited an acute Clsecretion response. In addition, short circuit current response in CF cultures exposed to UTP compared to normal cultures was markedly inhibited. Increases in forskolin or UTP dependent alveolar surface liquid height in normal alveolar cultures was significantly inhibited in the presence of the CFTR inhibitor or in primary CF alveolar type II cultures, suggesting a significant role for CFTR-mediated Clsecretion within alveolar type II cells. More importantly, alveolar wall liquid height was significantly increased, up to 24 h, using normal alveolar type II cultures subjected to cyclic compressive stress (CCS), a model used to mimic normal tidal breathing. In addition, we generated a transgenic mouse model expressing the secreted Clara Cell 10 (CC10) protein fused to eGFP driven by the mouse alveolar type II cell promoter, surfactant protein-C (SP-C). To test whether the transgenic secreted protein was transported into the upper airway, the trachea was isolated and lavaged (partial lavage). The secreted transgene was detected within the partial lavage of mice expressing the eGFP fusion protein. Collectively, our data uncover a significant novel CFTR-dependent secretory capacity within primary alveolar type II cells. Understanding the physiology of ion transport within the alveoli and the potential communication between the alveolar space and upper airways within physiological and pathological environments may be crucial to developing targeted therapeutics. The relative roles of CFTR and Ca 2+ -activated Clchannels (CaCC) in the function of airway submucosal glands is controversial. To address this issue we produced primary cultures of epithelial cells from human airway glands that formed tight epithelia when grown on permeable supports. In terms of their ultrastructure (granule size and opacity) and secretory products (lactoferrin, mucin) we established cultures that resembled either mucous or serous phenotypes by growth factor modifications and selection of support membranes. Serous cell differentiation was established on polyester (PET) filters in laminin-supplemented medium, and mucous cell differentiation on polycarbonate filters in EGF-containing medium. In Ussing chambers, basal Clcurrents were larger in mucous than serous cells, and were stimulated in both cell types by ATP or methacholine. With normal Cl --containing media on both sides, forskolin had little effect on Clcurrents, though it became markedly stimulatory in the presence of a serosal-tomucosal Clgradient. Baseline and mediator-induced increases in Clcurrents were inhibited by 40-60% by CFTRinh-172. To measure Clconductances directly we performed whole cell patch clamp studies of single isolated cells, which visually maintained their serous or mucous characteristics. The CFTR-mediated Clconductances in the presence of maximal intracellular cAMP concentrations was very small in the two cell types (serous, 50.6 ± 12.7 pS/pF; mucous 16.3 ± 4.1 pS/pF), but inhibited by ~66% by GlyH-101 and thus distinct from non-specific leak currents. ATP-stimulated conductances were also small. However, the maximal Clconductance induced by ionomycin was an order of magnitude larger (serous, 651 ± 73 pS/pF; mucous, 416 ± 152 pS/pF). Expression of mRNA for TMEM16a (the likely molecular basis for CaCC) was 26x (serous) and 104x (mucous) more abundant than mRNA for CFTR.

We have generated primary human gland cell cultures with serous and mucous ultrastructural and secretory features. 2) Serous and mucous gland cells express CaCC both molecularly and functionally at much higher levels than CFTR. 3) Ion transport characteristics of serous and mucous cell cultures were similar. 4) The relative contribution of CaCC to total Clcurrents is much greater when measured by patch-clamping of isolated cells than when determined from measurement of short circuit current in Ussing chambers. Supported by CF Research Inc. and CF Foundation. The apical HVCN1 H + channel of the airway epithelium contributes to cellular acid release into the airway surface liquid and likely supports the antibacterial activity provided by DUOX of the airways (Fischer H et al. Antiox Redox Signal 2009 Apr9 [Epub ahead of print]). A critical characteristic of the HVCN1 H + channel is its activation by an inside-to-outside H + gradient or by mucosal alkalinity, which is thought to be generated by bicarbonate secretion across CFTR. To investigate the regulation of HVCN1 by pH, we cloned the open reading frame of HVCN1 from total RNA prepared from human primary airway epithelial cultures from two donor lungs. In one lung we found two sequential polymorphisms at positions 452T->C and 453G->A resulting in a novel missense mutation M91T HVCN1. To determine the distribution of M91T HVCN1 in the human population, we investigated the genomic DNA Coriell panel SNP500V by PCR and MALDI-TOF mass spectrometry. In 95 analyzed samples, the M91T mutation was found in one heterozygous sample, which was further verified by sequencing. Wildtype and M91T HVCN1 were recombinantly expressed in Cos-7 cells as GFP fusion proteins and H + currents were investigated by whole cell patch clamping. We found that M91T HVCN1 resulted in H + currents that required significantly larger H + gradients than wildtype for the activation of H + currents. The relation of the activation potential to outside pH was shifted by 0.6 pH units to more alkaline values for M91T HVCN1. Further, we performed measurements of acid secretion by confluent airway epithelial cultures that natively expressed the M91T mutation (or wildtype) using the pH stat technique. HVCN1-dependent H + secretion was identified by its sensitivity to mucosal Zn 2+ (10 µM). The mucosal threshold pH above which HVCN1-mediated H + secretion was activated was found to be on average pH=6.9 in wildtype cultures and pH=7.5 in M91T cultures, indicating that the mutant is active only at quite alkaline pH values. In both whole cell and transepithelial measurements, the maximal H + currents achieved during large pH gradients were not affected by the mutation suggesting that the M91T causes a defect in the pH-dependent gating of HVCN1. We predict that this HVCN1 mutation will negatively impact airway disease if present in a CF subject.

Supported by CFF (FISCHE07G0), NIH (R21 HL089196, R01 HL086323). The underlying cause of increased mucus accumulation in cystic fibrosis (CF) affected organs is not understood. The facts that 1) mucus and HCO 3 are secreted concurrently in tissues in general, 2) HCO 3 secretion is decreased in CF-affected organs, and 3) the severity of clinical manifestations in CF appear to be associated with the gravity of cystic fibrosis transmembrane conductance regulator (CFTR)-dependent bicarbonate transport defect all suggest that HCO 3 must be critical to the process of normal mucus secretion. We previously found that the absence of serosal HCO 3 appears to inhibit mucus release. In the present study, we asked whether luminal HCO 3 -would support mucus release. We also examined the molecular components of HCO 3 secretion that may be crucial to the mucus release process. Methods: We perfused paired distal small intestinal segments of CL57BL/6 mice stimulated with prostaglandin E 2 or 5-hydroxytryptamine in the presence or absence of luminal or serosal HCO 3 and collected the perfusates to measure the amount of mucus released. In addition, we studied the response of mouse intestine carrying the most common CFTR mutation, ∆F508, to stimulation in the presence or absence of serosal .

We found that whereas the absence of serosal HCO 3 - inhibited mucus release by about 50% in wild-type intestine, the presence of luminal HCO 3 did not enhance mucus release. Blocking HCO 3 entry via inhibition of the Na + -HCO 3 cotransporter (NBC) with 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid (DIDS) decreased mucus release as much as did removal of serosal HCO 3 -, but carbonic anhydrase inhibition with acetazolamide did not decrease mucus release. Interestingly, the ∆F508 intestine showed poor or no response to mucus stimulation regardless of whether serosal HCO 3 was present or absent. Conclusions: Thus, concurrent HCO 3 secretion appears to be a necessary component of the mucus release process. The poor response of mucus release to luminal HCO 3 suggests that HCO 3 secretion must accompany, and must be in close proximity to the site of, mucus release. Functional NBC and CFTR appear to be essential components of mucus secretion. We propose that the characteristically aggregated mucus, observed in mucin secreting organs, is a consequence of defective HCO 3 transport in CF.

Ehre, C.; Button, B.; Randell, S.H.; Boucher, R.C. UNC-Chapel Hill, Chapel Hill, NC, USA Mucins are high-molecular-weight glycoconjugates that line the airway surfaces and protect the lungs against pathogens by means of their viscoelastic properties. Alterations of the macromolecular composition and the biophysical properties of the mucus gel affect its rheology, causing reduced transportability and adherence to cell surfaces in disease. In CF, mucus clearance is impaired, which leads to mucus plugging, chronic infection, chronic inflammation and subsequent irreversible tissue damage. Despite an increasing interest in mucin biology, the causes of airway mucus accumulation are still unknown. In this work, we present data comparing normal and CF specimen that show mucin gene expression, mucin composition of the plugs and introduce a novel assay to measure mucus adhesion. Fifteen disease-control and 5 CF lung samples were collected for this project at the time of lung transplantation or death. To ensure high-quality mRNA, tissue dissection took place immediately upon receipt. Specific regions of the lungs, i.e. segmental bronchi, intermediate-size and small (≤ 2mm diameter) airways were excised and either fixed or processed for mRNA analysis. The remaining lung tissue was processed to generate tracheobronchial air-liquid interface cell cultures. Following mRNA extraction, the levels of MUC5AC and MUC5B gene expression were measured by quantitative PCR and normalized to 18S. We showed that MUC5AC and MUC5B genes were upregulated in all-size CF airways, reaching up to 12.4-and 6.7-fold increase, for the respective mucins, in CF bronchi as compared to non-CF (p<0.05). Despite the lack of submucosal glands in the noncartilagenous airways, MUC5B expression was increased 5.9 fold in the small CF airways whereas MUC5AC expression remained low. Histology sections were used to confirm the absence of submucosal glands in the small airway samples. Immunostaining using new polyclonal antibodies against MUC5AC and MUC5B and AB/PAS staining revealed the presence of mucous glycoconjugates consistently in the airway surfaces of ventilated non-CF subjects but no mucus plug; whereas severe mucus obstruction occurred in CF airways. MUC5B signal was mainly detected in submucosal glands with few positive epithelial cells, while MUC5AC signal was restricted to the cell surfaces. Both mucins participated in the formation of the plugs. More work has to be done to determine the mucin subtype ratio in different areas of the lungs. Mucus adhesion in CF may be a crucial issue that requires to be addressed to restore mucus flow either by cough or by mucociliary clearance. Recently, our laboratory has generated a unique peel-test assay using a transducer device to measure the force required to remove mucus from epithelial surfaces and was used on normal and CF cultures and on organotypic lung explants. Our preliminary data showed increased mucus adhesion force in CF (1.4 mN) vs. normal (0.8 mN) cultures and measured a more robust force in organotypic explants (9.9 mN for normals, no CF was tested yet). This system offers a valuable in vitro model to uncover the causes of mucus adhesion and test pharmacological reagents, i.e. osmotic, emulsifying, chelators and/or mucolytic drugs, aimed to restore mucociliary clearance.

Supported by the CFF (MCC consortium). The airway surface liquid (ASL) is the thin fluid layer lining the airways whose depth is thought to be reduced in cystic fibrosis (CF). Prior measurements of ASL depth have been made in airway epithelial cell cultures. Here, we established methodology to measure ASL depth to ~1-µm accuracy in intact fragments of freshly obtained human and pig tracheas. Airway fragments were mounted in chambers designed for perfusion of the basal surface, and observation of the apical, fluorescently stained ASL by z-scanning confocal microscopy using a high numerical aperture lens immersed in perfluorocarbon. Analysis routines were developed to deduce ASL depths for every x,y-pixel in z-image stacks, in order to report data in an unbiased manner as histogram distributions of ASL depths. Average ASL depth in well-differentiated primary cultures of non-CF human nasal respiratory epithelium was 8.3 ± 0.8 µm (S.E., 8 cultures) under basal conditions, 8.0 ± 1.5 µm following ENaC inhibition by amiloride, and 14.5 ± 1.1 µm following CFTR stimulation by cAMP agonists. In CF nasal epithelial cell cultures, ASL depth was 6.3 ± 0.3 µm under basal conditions, slightly lower than that in the non-CF cultures, and not affected by IBMX/forskolin. ASL depth in human trachea was 7.0 ± 0.7 µm under basal conditions, 11.0 ± 1.7 µm following amiloride, 17 ± 3 µm following cAMP agonists, and 7.1 ± 0.5 µm after CFTR inhibition. Similar results were found in pig trachea. This study provides the first direct measurements of ASL depth in ex vivo fragments of intact human and pig airways, and indicates the involvement of ENaC sodium channels and CFTR chloride channels in determining ASL depth. We propose that a primary cause of CF lung disease is the inability of CFTR-deficient airways to increase their ASL depth following various secretory stimuli that in non-CF airways produce transient increases in ASL depth.

Hayes, E.F. 1 ; Bergin, D.A. 1 ; Vega-Carrascal, I. 1 ; Keenan, J. 2 ; Clynes, M. 2 ; Reeves, E.P. 1 ; O`Neill, S.J. 1 ; McElvaney, N.G. 1 1. Respiratory Research Division, Royal college of Surgeons, Dublin, Ireland; 2. National Institute of Cellular Biotechnology, Dublin City University, Dublin, Ireland Introduction: Cystic fibrosis (CF) is a life-threatening disease and respiratory infections are the leading cause of death. Paradoxically, phagocytic neutrophils are recruited into the lungs but fail to clear infections. The question that this project set out to address was; are CF neutrophils intrinsically abnormal? This study employed proteomic techniques that represent a valuable tool to identify global alterations in the protein expression profile of cells.

Aim: The objective of this investigation was to study alterations to the CF neutrophil membrane by gel-based two-dimensional electrophoresis (2DE) technologies to increase our understanding of the pathophysiological significance of the CF neutrophil and to identify putative markers with possible prognostic or diagnostic value.

Methods: Neutrophil membranes were prepared by sucrose-density ultracentrifugation. The solubilizing power of nonionic and zwitterionic detergents (2%SDS, 4% CHAPS, 2% Brij 56, 2% Triton X-100, 2% Tween 40, 2% Mega-10, 2% SB 3-12) as membrane protein solubilizers for twodimensional electrophoresis was investigated. For 2-D fluorescence difference gel electrophoresis (DIGE) proteins were minimally labeled with 400 pmol/50 µg of protein with either Cy3 or Cy5 for comparison on the same gel. Immobiline DryStrips pH 3-10 or 4-7 were employed. Gels were scanned using the Typhoon 9400 variable mode imager and analyzed using the DeCyder™ Software version 6.2. Statistical analysis and quantification of protein abundance was determined using the biological variation analysis module (BVA) of DeCyder™. Five biological repeats were used for neutrophil membranes obtained from normal and CF individuals.

Results: Optimized solubilization of membrane proteins was achieved by combining the zwitterionic detergent CHAPS (4%) or SB3-12 (2%) with the nonionic detergent Triton X-100 (2%). Comparison of membrane proteins of normal and CF individuals by 2-D DIGE revealed over 200 protein spots to be differentially regulated. Differential expression was defined as greater than 1.2-fold change in expression with a Student t test (p-value) of 0.05 or less.

Conclusion: There is significant evidence that the CF neutrophil is intrinsically abnormal. Data arising from this project will identify candidate proteins that could be used as biomarkers and/or contribute to a better understanding of disease progression in CF.

Supported by the Health Research Board Ireland.

Hadjiliadis, D. 1 Background: This is the first study to systematically evaluate lung function (spirometry and impulse oscillometry [IOS] ), CXCR2 agonists, and inflammatory biomarkers in the serum of stable outpatients with cystic fibrosis (CF) compared to healthy age-, gender, and racially-matched nonsmoking controls. Population: CF (n= 28); controls (n=14); mean age 33±11 years (range 20-55), 57% female, with an average FEV1 of 2.37±0.81L.

Results: Univariate analyses revealed higher CXCL8 (IL-8) and fibrinogen levels in CF patients. CXCL8 in CF: 10.2 pg/mL (CV: 66.1%) vs. controls: 5.9 pg/ml (CV: 81.6%) ratio 1.74 (CI 1.14, 2.66) (p= 0.0065); fibrinogen in CF: 3.93 g/L (CV: 21.9%) vs. controls: 3.00 (CV 13.6%) ratio 1.31 (CI 1.16, 1.49) (p=0.0004). There was a 74% increase in mean serum CXCL8 in stable CF outpatients compared to healthy age-and gender matched controls. Differences in serum CXCL1, MMP-9 and lung-derived biomarkers SP-D, and CC-16 were also observed but did not achieve statistical significance. No differences were observed for IL1β, CXCL 3, 5 and 7. Spirometry and effort independent IOS demonstrated significant differences between groups. There were no observed age or gender differences in inflammatory markers.

Conclusions: This study provides further support for testing the hypothesis that a CXCR2 antagonist will reduce the recruitment of neutrophils and hopefully prevent airway damage in CF. In addition, subject matching may not be necessary in future CF trials assessing these biomarkers.

Study funded by GlaxoSmithKline.

Background: Cystic fibrosis (CF) is a genetic multisystem disease leading to early mortality primarily due to respiratory failure. Lung function, weight and number of acute exacerbations per year are the three best predictors of death in CF and are commonly used surrogates for survivorship in clinical trials. Airway inflammation underlies declining lung function, and biomarkers of active inflammation may predict clinical outcomes and allow early detection and monitoring of progressive disease prior to irreversible lung damage. Such biomarkers may facilitate exploration of disease mechanisms and identification of new treatment targets.

Objective: To discover inflammatory markers predictive of clinically relevant outcomes.

Methods: We measured levels of high mobility group box-1 (HMGB-1), a potent novel mediator of sepsis and acute lung injury, tumor necrosis factor α, interleukins 1β, 6, and 8, myeloperoxidase and total sputum cell counts in sputum in adult patients with CF and adult volunteers without lung disease under IRB supervision. We looked for correlation with lung function, weight and number of acute exacerbations of disease.

Results: We collected sputum from 71 consecutively enrolled adult patients with CF and induced sputum from 15 volunteers. We found that a third of patients were in each of three categories: 1) stable, 2) suffering increased symptoms and 3) suffering increased symptoms with objective findings such as a significant acute drop in lung function. Each biomarker was substantially elevated in sputum from CF patients compared to normal volunteers. Among the biomarkers measured, HMGB-1 was the best predictor of decreased forced expiratory volume in one second (correlation coefficient -0.29, Kendall's τ, p < 0.0001), and was the only significant predictor of decreased body-mass index (-0.34, p < 0.001) and increased acute pulmonary exacerbations (5.2 x 10-4, ANOVA p < 0.001). In multivariate analysis, HMGB-1 was superior to all other biomarkers, singly and in combination, as a predictor for each of the three most critical endpoints predictive of survival in CF.

Conclusions: HMGB-1 is a clinically relevant biomarker of disease in CF. Measuring HMGB-1 may improve disease monitoring, refine estimates of individual prognosis and provide an additional endpoint uniquely informative of inflammation by which to evaluate new treatments. Should it prove to mediate CF progression, as it is does sepsis and acute lung injury, HMGB-1 may be an attractive target for pharmacological intervention.

Joo, N. 1 ; Cho, H. 1 ; Khansaheb, M. 1 ; Karp, P.H. 2 ; Stoltz, D.A. 2 ; Meyerholz, D.K. 3 ; Welsh, M.J. 2 ; Wine, J.J. 1 1. CF Research Laboratory, Stanford University, Stanford, CA, USA; 2. Internal Medicine, Iowa University, Iowa City, IA, USA; 3. Pathology, University of Iowa, Iowa City, IA, USA Human CF airway submucosal glands show a profound reduction in mucus secretion (Joo et al. J Biol Chem 2002; 277:50710; Salinas et al. Faseb J 2005; 19:431; Joo et al. J Biol Chem 2006; 281:7392; Song et al. Am J Physiol Cell Physiol 2006; 290:C741; Choi et al. J Clin Invest 2007; 117:3118; Choi et al. J Clin Invest 2009; 119:1189) . To evaluate the role of CFTR in airway gland secretion of the CF pig model (Rogers et al. J Clin Invest 2008; 118:1571; Rogers et al. Science 2008; 321:5897) , we used in situ optical monitoring to measure individual airway gland secretion rates in isolated tracheal mucosa from WT, heterozygous (cftr +/-, HZ) and CF (∆F508/null and null/null) piglets less than a week old. All secretion rates are expressed as picoliters/min/gland, followed by the number of glands and number of piglets tested (n). Average basal secretion rates were <1 in all genotypes (30-60 glands, n=3-5) . Mucus secretion rates for glands stimulated with 3 µM forskolin (Fsk), were WT: 73.7 ± 38.2; 38 glands, n=3, HZ: 68.0 ± 22.1; 46 glands, n=4, and CF: 1.3 ± 1.6; 30 glands, n=3. To assess the effect of CF mutations on synergistic gland secretion, we combined 100 nM carbachol (Carb), which alone induced no mucus secretion, with 3 µM Fsk. The combined treatment potentiated secretion rates in all genotypes: WT: 168.0 ± 29.7; 22 glands n=2, HZ: 230.0 ± 91.9; 34 glands, n=3, and CF: 36.0 ± 22.2; 30 glands, n=3. Substance-P (Subs-P) is a potent stimulus of gland secretion in pigs (Trout et al. Am J Physiol Lung Cell Mol Physiol 2001; 281:639) . Secretion rates to 1 µM Subs-P were WT: 212.5 ± 16.3; 22 glands, n=2, HZ: 215.0 ± 140; 14 glands, n=2, and CF: 52.7 ± 28; 24 glands, n=3. A high concentration of Carb (1 µM) strongly stimulated gland secretion: WT: 1895.5 ± 1637; 10 glands, n=2, HZ: 815 ± 345.8; 10 glands, n=2, and CF: 254 ± 116.3; 12 glands, n=3. The results demonstrate that mucus secretion from CF submucosal glands of piglets is greatly reduced in response to all agonists tested (forskolin, forskolin + low dose carbachol, Substance P, high dose carbachol) compared with secretion from WT and HZ controls of the similar age (CF piglets were 1 day old, WT: 3.1 ± 1.2, and HZ: 3.5 ± 1.7 days old). As a percent of the combined WT and HZ data, CF responses to each of the mediators were as follows: Fsk: ~2%, Fsk + low Carb: ~18%, Substance P: ~25% and high dose carbachol: ~19%. Our results from CF piglets suggest that CFTR plays a role in mucus secretion from pig airway glands that is similar in importance to its role in human airway glands. Because the CF piglets were tested within one day of birth, these differences predate infection. If CF pigs develop lung infections, the results will be consistent with the interpretation that defects in anion-mediated fluid secretion are part of the causal pathway leading to CF lung disease.

Supported Pediatric Pulmonology, Univ. of North Carolina Hospitals, Chapel Hill, NC, USA; 2. Cystic Fibrosis Center, Univ. of North Carolina Hospitals, Chapel Hill, NC, USA Background: In cystic fibrosis (CF), dysregulated ion transport is postulated to cause dehydration of the airway surface liquid (ASL) layer. ASL dehydration may cause impaired mucociliary clearance, thus predisposing to chronic endobronchial infections and an intense inflammatory response. We hypothesized that ASL dehydration represents the initiating event in CF airways disease; therefore, should be demonstrable early in life with appropriate measurement techniques. To test this, we assessed the status of ASL (percent solids) across a spectrum of CF and other diseases, extending into early childhood.

Methods: CF and disease controls (e.g. asthma, primary ciliary dyskinesia) <3 years of age and between 5-12 years of age were recruited at the time of a clinically indicated bronchoscopy (performed for exacerbation symptoms in CF). Another group of subjects 5-12 years of age were recruited for induced sputum during a baseline period. A novel method using a stationary cytology brush placed against the lateral wall of a mainstem bronchus during bronchoscopy was used to collect mucus that was transported via mucociliary clearance onto the brush over 3 minutes. The endobronchial brush (EB) samples were serially weighed with an ultramicrobalance over a 2 minute period to extrapolate a "time=0" wet weight. The brush was then dried overnight at 80∞C and re-weighed to determine a dry weight. Vigorous cleaning of the brush (in 6 M guanidine, using sonication) and redrying allowed subtraction of the "clean brush weight" so that percent (%) solids could be calculated. Induced sputum (IS) samples were analyzed using an established pre-post desiccation weight/gravimetric method to determine % solids.

Results: In CF subjects <3 years of age (n=2), EB mean % solids was 4.14%. In CF subjects 5-12 years of age (n=10), EB mean % solids was 6.49(+/-2.3)%. To date, a single disease control subject in the 5-12 year age group was noted to have an EB % solid value of 2.28%. IS samples from CF subjects in the 5-12 year old age group (n=4) had a mean % solid of 4.16(+/-1.69)%. IS samples in the disease control group (n=3), revealed mean % solids of 1.53(+/-1.15)%.

Conclusion: In this ongoing study, preliminary data reveals evidence for significant ASL dehydration in CF, using both induced sputum collection and a novel bronchoscopic method. Evidence for ASL dehydration was seen in the youngest subjects enrolled (<3 years of age), supporting the hypothesis that this is an early pathogenic event. For comparison, identical IS procedures in stable adults with CF (n=9) (unpublished data) revealed similar levels of dehydration (4.98+/-1.52%) whereas healthy adult volunteers (n=29) yielded lower % solids (1.93+/-2.05%). Also, adults with mild-moderate chronic obstructive pulmonary disease had less-pronounced ASL dehydration (2.8%+/-1.12%). Further studies are underway that will elucidate the nature of mucus constituents (i.e. mucin concentration, cell counts) and levels of potential regulators of hydration (i.e. nucleotides, cytokines).

Supported by NIH SCCOR grant 1P50HL084934; CF Foundation clinical fellowship grant; M01RR00046 and/or UL1RR025747 (NCRR).

Esther, C.R. 1 ; Olsen, B.M. 1 ; Boucher, R.C. 2 1. Pediatric Pulmonology, UNC Chapel Hill, Chapel Hill, NC, USA; 2. CF Research Center, UNC Chapel Hill, Chapel Hill, NC, USA Rationale: Analysis of exhaled breath condensate (EBC) offers an attractive, non-invasive, but as yet unproven method to track airway biomarkers in cystic fibrosis (CF). We report preliminary results from an ongoing longitudinal study designed to evaluate EBC purines and other small molecule biomarkers relative to lung function, quality of life, and sputum inflammatory markers in 50 children with CF.

Methods: The preliminary analysis included 30 EBC samples obtained from 14 subjects; 9 subjects had two or more EBC samples for longitudinal evaluation. Mass spectrometric methods were utilized to simultaneously measure EBC concentrations of previously identified small molecule biomarkers including purines (adenosine, AMP), amino acids (phenylalanine, tyrosine, valine), and others (sialic acid, taurine) plus urea as a dilution marker. Biomarkers were expressed as ratios to urea to control for dilutional variability of airway secretions in EBC.

Results: Purines, amino acids, and urea were detected in all EBC samples; taurine was detected in only 67% and sialic acid in only 47%. Among the biomarkers, adenosine was most closely linked with longitudinal changes in lung function, with a negative relationship observed between EBC adenosine/urea and percent predicted FEV 1 within individual subjects (mean slope by linear regression -0.026±0.020, p<0.01, Figure) . EBC adenosine/urea was also elevated when percent predicted FEV 1 fell at least 5 percentage points below baseline (0.52±0.29 vs. 0.29±0.12, p<0.01), and there was a correlation trend between average EBC adenosine/urea and average percent predicted FEV 1 (r=-0.46, p<0.10). Similar observations for AMP/urea did not reach statistical significance. No relationships between lung function and other EBC biomarkers were observed. Relationships between EBC biomarkers and quality of life or sputum inflammatory markers await analysis of a larger sample set.

Conclusions: EBC adenosine is a promising biomarker that appears to track changes in lung function in CF, consistent with its role as a biomarker in other lung diseases. Further analyses of the longitudinal study are ongoing to examine these conclusions.

This study was funded by CFFT.

Linear regression of the relationship between EBC adenosine/urea and lung function in individual subjects. Previous work from our laboratory has implicated the EP 4 . prostanoid receptor as playing an essential role in CFTR-mediated transepithelial anion secretion across, and anion efflux from, the human airway epithelial cell line, Calu-3 in response to the isoprostane molecule 8-iso-PGE 2 . Isoprostanes are produced when oxygen-centred free radicals react with membrane bound polyunsaturated fatty acids and their levels are elevated in exhaled breath condensates from CF patients, indicating they may be important mediators of oxidant stress in the airways. Recently, levels of PGE 2 measured in exhaled breath condensates were also found to be significantly higher in CF patients compared to controls (Lucidi et al. Free Radic Biol Med 2008; 45:913-9) . In order to more thoroughly understand the physiological role of the EP 4 receptor in airway epithelial cells, we investigated the signalling pathways and downstream events initiated by activation of this receptor. Treatment with either the isoprostane 8-iso-PGE 2 (1 µM or 3 µM) or the EP 4 receptor selective agonist PGE1-OH (1 µM or 3 µM) resulted in the phosphorylation of two major ERK pathway proteins, MEK1/2 and p44/42MAPK in the cytoplasm, whereas there was no detectable change in phosphorylation status of c-Raf, p90RSK or Elk1. We additionally investigated whether these agonists induced activation of the transcriptional factor early growth response (Egr-1), which regulates downstream target gene expression via binding to an Egr nuclear response element. Using Western blotting, we determined that neither agonist (at 1 µM) induced transient induction of early growth response Egr-1 protein. However, PGE1-OH at 3 µM showed a significant increase of Egr-1 protein expression, detectable in the nuclear fraction as soon as 30 min post-treatment. This increase of Egr-1 protein expression was abolished in the presence of the EP 4 receptor antagonist AH23848, confirming it is an EP 4 receptor mediated event. In addition, we investigated whether EP 4 receptor activation ultimately resulted in the additional production of PGE 2 , which could potentially act either in a positive-feedback way via the EP 4 receptor, or alternately, act at other EP prostanoid receptor subtypes in either an autocrine or paracrine manner. Using misoprostol (which stimulates EP 2 , EP 3 and EP 4 receptors) as our agonist (at 1 µM for 24 hours), we determined that treatment of Calu-3 cells was associated with an increased production of PGE 2 as measured by ELISA.

In conclusion, activation of the EP 4 prostanoid receptor, which could occur either by isoprostanes produced as a consequence of oxidative stress or via endogenous PGE 2 , results in activation of the MEK1/2 and p44/42MAPK pathways, and enhanced PGE 2 production. Given the pleiotropic effects of PGE 2 in the airways, EP 4 receptor activation may play an important role in airway physiology via increased production of PGE 2 .

Supported by the Canadian Cystic Fibrosis Foundation. Background: Several studies report that mucociliary clearance (MCC) is impaired in adults with cystic fibrosis (CF). Because MCC is an important airway defense mechanism, drugs that slow impairment of MCC in children could prove beneficial in the long-term prognosis of the disease. However, little is known about MCC in children with CF. This study was designed to determine basal MCC values in these children and the MCC response to acute inhalation of 7% hypertonic saline (HS), an osmotic stimulus.

Methods: Twelve children with CF (7-12 yrs; 5 males) and normal pulmonary function (FEV1 and FVC >90% of predicted values) inhaled 0.12% saline (placebo) and HS in a double-blind, randomized, cross-over study. Following inhalation of saline, patients inhaled the radioisotope 99m-technetium and underwent sequential imaging of their lungs with a gamma camera for 90 min and approximately 24 hrs later. Mucociliary clearance was quantified at 20 min (MCC20), 60 min (MCC60), 90 min (MCC90) and 24 hrs (MCC 24hrs) after inhalation of the radioisotope. Between the 60 min and 90 min measurements, children coughed 30 times. Measurements of MCC following inhalation of 0.12% saline in the children were compared to MCC measurements obtained in nine healthy adult control subjects (18-37 yrs). Measurements of MCC on the two treatment visits in the children were also compared.

Results: Mean MCC60 (± standard deviation), considered an index of large-airway clearance, was similar for the children treated with 0.12% saline and the adult controls, with 18.1±8.6% and 16.4±11.3%, respectively. However, MCC60 in the five male children with CF averaged 11.8±5.5% and this was statistically lower than MCC in the seven female children with 22.7±6.7% (p = 0.02). MCC 24hrs, considered an index of large and small airway clearance combined, was reduced in the children with CF, averaging 21.1±13.7%. However, this was not statistically different from MCC 24hrs measured in the controls, which averaged 33.6±16.3%. MCC20 was similar in the children after treatment with 0.12% saline and HS, averaging 11.6±6.7% and 9.5±6.9%, respectively. MCC60 was higher in seven of the twelve children after HS, but the mean value (19.7±9.3%) was not statistically different from the mean MCC60 with 0.12% saline (18.1±8.6%). Mean MCC90 and MCC 24hrs increased after both treatments, but neither treatment was statistically different from the other at these two time points.

Conclusions: These results suggest that basal MCC from the large and small airways of this group of 7-12 yr old children with CF and normal pulmonary function is similar to that of healthy, adult controls and is not yet impaired. However, basal MCC in male children in this group appears to be lower than female children. More than half of the children showed improvement in MCC after treatment with HS. The clinical impact of improvement in MCC with HS in these children is unknown. The objective of this study was to investigate the role of CFTR in airway epithelial cell migration using Calu-3 and normal human bronchial epithelial (NHBE) cells. Migration was assessed by measuring changes in impedance (Z) at the membrane-substrate interface as cells migrated onto the surface of an electrode to close a wound in the epithelium. Inhibition of CFTR activity with CFTR inh -172 slowed the rate of wound closure (time to reestablish initial Z) of Calu-3 cells (30 µM) by 6.2 hours compared to control cells (serum free media, 1µg/ml EGF), and of NHBE cells (20 µM) by 10.0 hours compared to control cells (serum free media). A CFTR specific effect on migration was verified in Calu-3 cells by constitutive expression of a CFTR specific shRNA which reduced CFTR mRNA expression by >99%. Silencing CFTR in Calu-3 cells delayed wound closure by 9.8 hours compared to wild type Calu-3 cells and cells expressing a non-specific shRNA. Silencing or inhibition of CFTR in Calu-3 cells reduced the area of lamellipodia projected from the leading edge of wounds by >2-fold, and CFTR inhibition in NHBE cells reduced this area by 2.6-fold. Replacement of Clions with in the media during Calu-3 cell migration had no effect on cell migration rate. However, reduction of [HCO 3 -] from 26 mM to 5 mM in HEPES buffered CO 2 gassed media slowed the rate of Calu-3 cell migration by 3.7 hours, in the absence of a change in intracellular pH. The results of these experiments indicate that functional activity of CFTR is involved in Calu-3 and NHBE cell migration and suggest a role for CFTR in airway epithelial restitution. Further, these results suggest that CFTR participates in HCO 3 transport which appears to be necessary for efficient cell migration.

Dysregulation of ion transport in lung epithelial cells is of critical importance in patients with cystic fibrosis (CF). Recent in vitro work has demonstrated that stimulation of the beta-2 adrenergic receptors (ADRB2) results in stabilization of chloride channels and restoration of secretory capacity in glandular serous cells that is dependent on CFTR. In addition, ADRB2 stimulation results in an increase in the number of epithelial Na + channels on type-I alveolar cells and increased transfer of Na + into the cells. Pursuant to planned studies in CF, we sought to determine the influence of an inhaled β-agonist on exhaled breath condensate (EBC) volume and Na + , K + , and Clin healthy subjects (n=13, age=28±9years, ht=178±10cm, wt=73±14kg, BMI=23±5kg/m 2 , VO2Peak=105±28%predicted, FVC=94±10%predicted, FEV1=95±15%predicted, mean±SD). To determine this, we collected EBC (for Na + , K + , and Cl -) for 25 minutes, along with serum Na + , K + and Cl -, before and 30, 60, and 90 minutes following the administration of a β-agonist (nebulized albuterol sulfate, 2.5mg diluted in 3ml of normal saline). The administration of a β-agonist resulted in a small decrease in serum K + and Clthat was significant (p<0.05) at 60min following albuterol administration but there was no change observed in serum Na + (Na + =139±1 and 139±1mmol/l, K + =4.1±0.5 and 3.9±0.4mmol/l, Cl -=105.5±1.5 and 104±2mmol/l for baseline and 60min following albuterol). The volume of EBC collected was reduced with albuterol and became significant (p<0.05) after 60min (4.0±0.6, 3.8±0.6, 3.7±0.5, and 3.7±4ml, for baseline, 30, 60 , and 90min post β-agonist). Exhaled Na + was lower following the β-agonist, whereas Clvalues increased, but there was no significant change in exhaled K + (Na + =3.4±1.9, 2.2±0.6, 2.3±0.7, and 2.4±1.3mmol/l; K + =0.028±0.02, 0.029±0.02, 0.022±0.01, and 0.021±0.01mmol/l; Cl -=69±41, 84±54, 96±36, and 70±27µmol/l, for baseline, 30, 60, and 90min post, p<0.05 for Na + and Cl -). These results demonstrate that administration of an inhaled β-agonist results in a decrease in exhaled Na + and in the volume of EBC along with an increase in exhaled Cl -. These findings confirm previous in vitro studies that have demonstrated ADRB2 stimulation enhances Clefflux and Na + influx from the airspace of the lungs and highlight the future importance of similar studies in patients with CF. In cystic fibrosis (CF) lungs salt and water transport of airway bronchial epithelium is impaired resulting in mucus adhesion and chronic bacterial colonization. We analyzed the influence of the cyclic AMP (cAMP), GlyH-101 and the potentiator UCCF-853 on paracellular fluorescein permeability of model bronchial epithelium 16HBE14o-and CFBE41o-. In 16HBE14o-cAMP produced a decrease of transepithelial electrical resistance (TER) by 50% and strong increase of paracellular permeability of the marker molecule fluorescein. The ∆F508-CFTR expressing CFBE41o-cells show under control conditions paracellular permeabilities similar to those of 16HBE14o-. Exposure of CFBE41o-cells to cAMP increased TER slightly without influencing the permeability of the paracellular pathway. Interestingly, the CFTR specific blocker GlyH-101 was without influence on the cAMP induced changes in paracellular permeability of 16HBE14o-cell layers. Also the CFTR potentiator UCCF-853 did not change the fluorescein flux of 16HBE14o-cells under control and stimulated conditions. However, UCCF-853 caused a significant increase of paracellular permeability in CFBE41o-cell layers upon cAMP stimulation. Temperature rescuing (27∞C for 48h) changes the situation completely. Now cAMP is able to increase the paracellular fluorescein flux in CFBE41o-cell layers considerably. This augmentation of fluorescein permeability is further increased by UCCF-853. After temperature rescuing the changes of paracellular permeabilities upon cAMP and UCCF-853 were comparable in CFBE41o-and 16HBE14o-cells. Of note, there was also no effect of GlyH-101 on the fluorescein flux in these temperature-rescued cell lines. It is worth mentioning that we observed no difference between basolateral to apical and apical to basolateral fluorescein flux.

Our data suggest strongly that CFTR is involved in the regulation of the paracellular permeability. This regulation presumably depends on the membrane presentation of CFTR and not on its chloride conductance. Together, our results link CFTR dysfunction to improper regulation of the paracellular transport and it is likely that this contributes to the symptoms of CF pulmonary phenotype. The small airways are a primary target in the pathogenesis of cystic fibrosis and other airway diseases. Although small airways account for the largest portion of the total surface area, characterization of the epithelial ion transport property functions of small airways has been limited by their relative inaccessibility, especially in native human tissue.

Methods: Recently, we introduced a novel method that provides significantly improved measures of electrolyte transport parameters in the small airways (≈1 mm i.d.) ex vivo in their native state. To rigorously define small airways ion transport properties, we used our recently developed micro-Ussing chamber (area=1 mm 2 ), to obtain the first measurements of ion transport properties of human native small airways specimens (2.6±0.3 mm dia.) of 12 donor subjects. Open circuit transepithelial potentials were recorded and constant current pulses (0.7-1.1 µA) were passed across dissected small airway tissue to determine transepithelial resistance.

Results: In bilateral 150 mM NaCl Ringer solution, spontaneous transepithelial potential (V t ), resistance (R t )(conductance (G t )) and equivalent short circuit current (I sc eq ) were: -4.1±0.5 mV, 66.0 ± 11.6 Ωcm 2 (20.1±2.9 mS/cm 2 ) and 78.1±14.0 µA/cm 2 (mean±SE) respectively. When 150 mM gluconate replaced luminal chloride, V t hyperpolarized markedly to -27.0±1.9 mV (n=13; p<0.001), R t increased to 105.3±19.9 Ωcm 2 (n=13; p<0.001); (G t decreased to 12.8±1.9 mS/cm 2 , n=13; p=0.0001).

In the presence of symmetric NaCl (150 mM NaCl in the bath and lumen) activating Clconductance with β-adrenergic (cAMP-mediated) agonists forskolin (Fsk, 10 µM) plus IBMX (100 µM) and purinergic agonist UTP (100 µM) evoked distinctly opposite effects on V t . That is, cAMP-mediated activation of Clconductance depolarized V t significantly, while apparently activating a purinergically mediated Clconductance with UTP, hyperpolarized V t significantly. Both agonists increased G t . Also luminal amiloride significantly depolarized the V t to -2.4±0.8 mV (n=8; p<0.001) and significantly decreased G t to 19.7±6.2 mS/cm 2 , (n=8; p=0.003). On the other hand, in the presence of a Clgradient, stimulating with Fsk plus IBMX and UTP hyperpolarized V t and decreased R t significantly. Furthermore, luminal amiloride depolarized V t to -18.1±1.2 mV (n=10; p<0.0005) and significantly increased R t to 111.4±26.3 Ω-cm 2 (n=10; p=0.002). These results suggest that, a) human small airway epithelia express very high constitutively active Clconductances that support both secretory as well as absorptive functions, b) cAMP may be most active in stimulating a group of cells with cAMP-sensitive Clchannels (CFTR) that absorb Cl -, while UTP may stimulate a group of cells with purinergically activated Clchannels (Ca 2+ -activated Clchannel) that secrete Cl -.

Conclusion: Small airways appear to consist of separate populations of secretory and absorptive cells that locally recycle airway surface fluids to maintain appropriate airway hydration. Relatively low capacities to secrete relative to absorb may compromise airway clearance in disease such as cystic fibrosis.

Supported by the Nancy Olmsted Trust, NIH (R01-HL084042) and American Lung Association California. Cytosolic phospholipase A2 (cPLA2) is a key enzyme controlling the release of arachidonic acid (AA) from membrane phospholipids. Conversion of AA by cyclooxygenases (COX) and lipoxygenases (LOX) generates prostaglandins and leukotrienes, respectively. We have previously observed that cPLA2 is involved in the mucus overproduction and mucin MUC5AC expression in CF mice. Our aim is to identify AA metabolites and signalling pathways involved in MUC5AC expression in the bronchial epithelial cell line NCI-H292. Measurements of MUC5AC expression by ELISA and qRT-PCR showed that PMA-induced MUC5AC production was mimicked by AA and inhibited by cPLA2 inhibitors. MUC5AC expression was inhibited by a general LOX inhibitor (nordihydroguaiaretic acid or NDGA) but not by COX inhibitors (aspirin, NS398). Inhibitors of 12-LOX (cinnamyl-3,4-dihydroxy-alpha-cyanocinnamate or CDC and baicalein), but not those of 5-LOX or 15-LOX, reduced MUC5AC expression. These inhibitors did not reduce IL-8 secretion. In agreement, 12-HETE, the first AA metabolite by 12-LOX, stimulated MUC5AC production. Baicalein reduced ERK activation and SP-1 translocation induced by PMA. In addition, 12-HETE activated ERK and SP-1 translocation. Both 12-R-LOX and 12-S-LOX forms are expressed in NCI-H292. Using siRNA targeting 12-R-LOX or 12-S-LOX we showed that 12-R-LOX rather than 12-S-LOX is involved in PMAinduced MUC5AC secretion. Our data suggest that 12-R-LOX pathway is involved in MUC5AC production via ERK and SP-1-dependent mechanisms. Potential therapeutic approaches targeting 12-LOX would be of great interest in the treatment of mucus overproduction in chronic lung diseases such us CF.

Work supported by the French CF association "Vaincre la Mucoviscidose." In some inherited disease states, including cystic fibrosis (CF), an abnormal control of cellular Ca 2+ homeostasis is observed. Here, we identified TRPC1 and TRPC6 isoforms of non-selective, calcium-permeable cation channels in human normal and CF airway epithelial cells. Although the activity of store operated TRPC1 Ca 2+ channels was similar in these cells, the application of the membrane permeant diacylglycerol analog, oleyl-acetyl glycerol (OAG) activated TRPC6-mediated Ca 2+ influx that was abnormally increased in CF (F508del/F508del) compared to non-CF cells. After correction of abnormal F508del-CFTR trafficking in CF cells, the level of OAG-dependent and TRPC6 mediated-Ca 2+ influx was also normalized. In CF cells, siRNA knockdown of endogenous TRPC6, but not TRPC1, reduced the abnormal OAG-dependent-Ca 2+ influx. Co-immunoprecipitation experiments revealed TRPC6/CFTR and TRPC6/F508delCFTR interactions in CF or non-CF epithelial cells. By contrast, we found no evidence for TRPC1/TRPC6, TRPC1/CFTR or TRPC1/F508delCFTR interactions. These data indicate that TRPC6 and CFTR are functionally coupled within a molecular complex and that CFTR down-regulates TRPC6 activity in non-CF airway epithelial cells. On the contrary, because this functional coupling is lost in CF cells, the OAGdependent-Ca 2+ influx is abnormal. In conclusion, we identified TRPC6 as the channel responsible for Ca 2+ influx in human CF cells linking in airway epithelial cells, the calcium signalling to CFTR pathobiology.

Supported by the French association "Vaincre La Mucoviscidose" and CNRS.

Lee, R.J.; Foskett, K. Physiology, University of Pennsylvania School of Medicine, Philadelphia, PA, USA Submucosal glands secrete a significant amount of the fluid and mucus that line cartilaginous airways, facilitating mucociliary clearance and likely playing a role in airway diseases such as cystic fibrosis (CF). Previous studies of intact glands have yielded insights into the regulation, volume, and composition of their secretions, but the contributions of the various glandular cell types (serous, mucous, collecting duct) remain poorly defined. The serous acinar cells, thought to be the primary source of secreted fluid, are major sites of CFTR expression in the lung. We previously developed optical methods to study fluid secretion in primary serous cells isolated from murine nasal glands. However, data has emerged suggesting that fundamental differences exist between murine and human glands, particularly concerning CFTR-dependent secretion. We have thus begun examining serous cell secretion in the pig, an animal model more similar to humans in terms of airway gland distribution and physiology. Intact submucosal glands were dissected from adult WT porcine bronchi and enzymatically digested. Serous acinar cells were identified based on visible morphology, immunocytochemistry, and gene expression profiling. Cells were studied using DIC imaging of cell volume to track agonist-induced changes in cell solute content in combination with simultaneous fluorescence microscopy of indicator dyes to track changes in the concentrations of ions involved in driving fluid secretion (Cl -) and regulating it (Ca 2+ ). Stimulation of serous cells with 10 µM carbachol (CCh; a cholinergic agonist) was associated with a profound elevation of intracellular [Ca 2+ ] ([Ca 2+ ] i ) from 50±6 nM at rest to a peak of 365±38 nM within 68±11 sec (n=10), accompanied by rapid cell shrinkage of 24±2 % within 39±15 sec after the onset of the [Ca 2+ ] i response. Elevation of [Ca 2+ ] i was found to be both necessary and sufficient to activate cell shrinkage. Upon return of [Ca 2+ ] i to resting levels, cells swelled back to resting volume (time to 50% volume recovery 150±28 sec; n=5). Agonist induced volume changes were associated with changes in [Cl -] i , suggesting that shrinkage and swelling are indicative of solute efflux and influx, respectively. The rate of CCh/Ca 2+ -evoked shrinkage/Clefflux was inhibited by the Ca 2+ -activated Clchannel inhibitor niflumic acid (50 µM), while the CFTR inhibitor GlyH-101 (10 µM) had no discernable effect, suggesting that Ca 2+ -evoked serous cell secretion is primarily CFTR-independent, as previously observed in murine cells. Cell swelling upon return of [Ca 2+ ] i to resting levels was inhibited by both the Na + K + 2Clcotransporter (NKCC) inhibitor bumetanide (100 µM) and the Na + H + exchanger (NHE) inhibitor dimethylamiloride (30 µM; times to 50% volume recovery were 640±145 sec [n=7] and 500±85 sec [n=5], respectively). Thus, both NKCC activity and paired NHE and Cl -/HCO 3 exchanger activity likely accumulate cell solute content to sustain transepithelial secretion. Together, these data suggest that agonist-induced volume changes reflect changes in the secretory state of porcine bronchial serous cells, and that cell volume imaging will be useful for identifying mechanisms involved in serous cell fluid secretion.

Lee, R.J.; Foskett, K. Physiology, University of Pennsylvania School of Medicine, Philadelphia, PA, USA

The serous acini of airway submucosal glands are thought to be the origin of the majority of the fluid secreted by these glands. Serous cells are sites of CFTR expression in the lung, and thus they may be critical to CF pathology. Fluid secretion is significantly reduced in intact airway glands from CF patients stimulated with the physiological secretagogue vasoactive intestinal peptide (VIP) or the adenylyl cyclase-activating compound forskolin, both of which are thought to signal through cAMP production leading to downstream activation of CFTR via PKA phosphorylation. We isolated serous acinar cells from adult WT porcine bronchi to (1) directly examine if these cells secrete fluid in response to VIP, (2) determine whether or not CFTR functions as a secretory Clchannel in these cells, and (3) identify other ion channels and transporters involved in serous cell secretion. Serous cells were studied using DIC imaging of cell volume to track changes in cell solute content along with simultaneous fluorescence microscopy of either fura-2 or SPQ to measure changes in intracellular ] i responses elicited by CCh (100nM and 1µM), suggesting cross-talk exists between cAMP and Ca 2+ signaling pathways in these cells. By measuring simultaneous changes in cell volume and SPQ fluorescence, cell volume changes during VIP stimulation were found to be associated with changes in [Cl -] i , suggesting that VIP-stimulated cell shrinkage reflects Clefflux as observed for CCh-stimulated shrinkage. The rate of VIP/forskolin-stimulated shrinkage/Clefflux was significantly inhibited by the CFTR-inhibitor GlyH-101 (10µM), while the Ca 2+ -activated Clchannel inhibitor niflumic acid (50µM, which slows the rate of CCh-stimulated shrinkage) had no effect. These data suggest that VIP-evoked Clefflux occurs at least partly through CFTR. Whole cell patch clamping is currently being used to examine/characterize both cAMP-and Ca 2+ -activated Cland K + conductances. The combination of optical and electrophysiological techniques will allow us to directly determine if VIP stimulation activates CFTR and whether or not the observed Ca 2+ signals are required for K + channel activation to mediate counter-ion current required for Clsecretion. Prior investigations suggest that, in contrast to the nasal mucosa, CFTR plays a minor role in murine tracheal epithelium. However, the activity of CFTR has not been examined in native distal airways, where the pathophysiology of human cystic fibrosis occurs. We sought to define CFTR-mediated chloride conductance in lower murine airway epithelium using a novel ex vivo microperfusion assay. Our hypothesis is that CFTR contributes a small chloride conductance in this tissue, similar to the trachea.

Wild type (WT) and ∆F508 CFTR mutant mice bred on a C57BL/6 background were used for all experiments. After deep anesthesia is induced using ketamine and xylazine, the left lung is surgically removed and dissected under cold conditions. The mainstem airway is mounted on concentric glass pipettes and immersed in a warmed bath of Ringer's solution. The airways are perfused with solutions of defined composition, including agonists and antagonists of CFTR conductance. Substituting luminal chloride with gluconate, a relatively impermeant anion, generates a transepithelial chloride gradient. The transepithelial potentials (TEP) are recorded in real time, including their deflection in response to various luminal perfusates.

Results: IBMX with forskolin hyperpolarizes the TEP in the presence of a chloride gradient, indicating a cAMP-sensitive chloride conductance. This conductance is blocked by CFTR inh -172. The WT airways exhibit significantly greater responses to both cAMP-mediated CFTR stimulation (∆mV -5.1 +/-0.5 for WT vs. -1.8 +/-0.9 for CF, p < 0.005) and inhibition (∆mV 4.8 +/-0.7 for WT vs. 2.2 +/-0.6 for CF, p < 0.05) than those from ∆F508 CFTR littermates. We also find a greater amiloride-sensitive potential generated by the CF tissue (8.8 mV +/-0.3 for WT vs. 11 mV +/-1.4 for CF, p < 0.05).

Discussion: We present an ex vivo microperfusion assay of the intact murine lung, developed to better replicate the native environment of the distal airways. Our data demonstrate small but significant CFTR activity in the distal airways of both wild type and CF mice. The reproducible response to IBMX with forskolin and CFTR inh -172, and its diminished effect in the ∆F508 mouse, suggest that this is a CFTR-mediated chloride conductance. These findings support those seen in murine tracheal epithelium, which exhibits predominantly calcium-activated chloride conductance with a seemingly decreased dependence on CFTR. The greater amiloride-induced depolarization generated in the CF mouse suggests augmented epithelial sodium channel activity or possibly lower, but undetected, CF airway chloride conductance. As the CF mouse does not exhibit airway pathology, it is unclear whether this plays a role in vivo. The apparent residual CFTR activity of the ∆F508 distal airways suggests persistent, albeit attenuated, expression, the role of which awaits further exploration. Recent studies have identified alterations in cholesterol processing associated with cystic fibrosis (CF). Some studies suggest that cholesterol processing alterations in CF are the result of deltaF508 CFTR trafficking deficiencies resulting in disrupted lipid trafficking. Previous observations demonstrate that Cftr-null cells and tissues exhibit an increase in plasma membrane cholesterol content and an increase in de novo cholesterol synthesis compared to wild type controls arguing against the role of deltaF508 CFTR trafficking as an initiating source of cholesterol accumulation. The hypothesis of this study is that these CF-related alterations in cholesterol processing will be influenced by Cftr genotype, but not dependent on the deltaF508 mutation. De novo cholesterol synthesis was assayed in mice homozygous for either the R117H (R/R) or the deltaF508 (F/F) CFTR mutations. Results reveal a 1.6 ± 0.2 fold (p < 0.01) increase in % new cholesterol synthesis in the lung of F/F mouse lung compared to controls and a 1.6 ± 0.3 fold (p = 0.04) increase in the liver. Similar results were found in the R/R mouse with a 1.2 ± 0.1 fold (p = 0.04) and 1.7 ± 0.2 fold (p = 0.008) increase respectively in the lung and liver. Membrane cholesterol content is determined by electrochemical detection and de novo synthesis is monitored by assessing deuterium-labeled water incorporation into cholesterol over an eight-hour period. In order to determine if membrane cholesterol content correlated with Cftr genotype, membrane cholesterol content was measured in nasal epithelium isolated from R/R and F/F mice. Membrane cholesterol content as measured by response ratio is elevated in both R/R and F/F nasal epithelium (1.64 +/-0.09 (R/R), p < 0.001; 2.14 +/-0.10 (F/F), p = 0.01) compared to age-matched sibling wt controls. These data demonstrate that either a mild or severe disease-related CFTR mutation will result in an increase in membrane cholesterol, with a larger magnitude increase in the F/F tissue. The magnitude increase in membrane cholesterol content in F/F mouse nasal tissue is identical to what is observed in Cftr -/-nasal tissue. The next goal was to determine if there is a CFTR dose effect with Cftr +/mice. Cftr +/-nasal epithelium exhibits a relatively slight, but significant loss of membrane cholesterol (0.87 +/-0.04 fold compared to wt, p < 0.01). These data suggest that CFTR function impacts cholesterol movement to the plasma membrane. To test this hypothesis directly, 9/HTEo-cells were treated with CFTRinh-172 and membrane cholesterol assayed. Contrary to cellular and in vivo CF models, acute CFTR inhibition resulted in a dramatic reduction in membrane cholesterol content. This finding suggested that increased membrane cholesterol content in CF is actually a secondary response. This study demonstrates that alterations in cholesterol synthesis and membrane cholesterol content are not dependent on deltaF508 CFTR trafficking disruptions and that secondary responses may be likely for increases in membrane cholesterol content in response to lost CFTR function. Results also suggest that membrane cholesterol content of nasal epithelium could prove to be an effective biomarker for CFTR correction.

This study is supported by funding from the CFF. 2008;118:1578) . Although widely used as a model for airway mucus secretion, no previous single-gland studies have been reported for fluid secretion from ferret glands. Therefore, we have carried out such studies in the hope that they can provide normative, parametric data to help inform studies of gland secretion in CF ferrets. We used in situ optical monitoring to measure individual airway gland secretion rates in isolated tracheal segments with intact cartilage from ferrets, aged 80 ± 4 days. All tracheas were obtained after acute experiments for other purposes. Secretion rates are expressed as picoliters/min/gland, followed by the number of glands and number of ferrets tested. Gland density based on actively secreting duct openings was 2-3 per mm 2 . Average basal secretion rates were < 3 (256 glands, n=13). Carbachol (1 µM) increased secretion rates to 1027.8 ± 309.8 averaged over the first 10 min (53, n=5) with early peak responses. Vasoactive intestinal peptide (VIP, 1 µM) and forskolin (10 µM, Fsk), which elevate cAMP, each produced sustained secretion of 186.7 ± 63.0 (57 glands, n=6) and 243. 3 ± 55.8 (34 glands, n=3) . The secretion pattern lacked the early peak seen with carbachol and required at least 30 min to reach a maximum rate that was then sustained. Ferret glands responded to Fsk (10 µM) with secretion rates that were 23.3 % of the secretion rates to 1 µM carbachol. The α-adrenergic agonist phenylephrine is a weak agonist for gland secretion in humans, sheep and pigs, but a strong agonist in cats (Joo et al. Am J Physiol Lung Cell Mol Physiol 2001; 281:458) . Ferrets responded like cats, with 10 µM phenylephrine producing a rate and pattern of secretion similar to that seen for carbachol. The overall and sustained secretion rates were 1655.4 ± 389.1 (62 glands, n=6) and 1721.3 ± 400.8 (59 glands, n=6); values slightly higher (118%) than those obtained for carbachol. Except for responsiveness to phenylephrine, ferret glands appear similar to human glands in their responses to mediators. It will be of great interest to determine how these responses are affected in CFTR -/-ferrets.

Supported by CFF and NIH.

CF nasal tissues have ion transport defects similar to those found in lower airways (Knowles et al. Science 1983; 221:1067) . Submucosal glands in CF airways hyposecrete mucus (Joo et al. J Biol Chem 2002; 277:50710; Jalinas et al. Faseb J 2005; 19:431; Joo et al. J Biol Chem 2006; 281:7392; Song et al. Am J Physiol Cell Physiol 2006; 290:C741; Choi et al. J Clin Invest 2007; 117:3118; Choi et al. J Clin Invest 2009; 119:1189) . To investigate a potential role of CFTR in submucosal gland secretion of the CF piglet nasal turbinate, we used in situ optical monitoring to measure individual airway gland secretion rates in isolated nasal turbinate mucosa from WT and CF (∆F508/null and null/null) piglets less than 3 weeks old. Secretion rates are expressed as picoliters/min/gland, followed by the number of glands and number of piglets tested (n). Average basal secretion rates were <5 in both genotypes (73 glands, n=7). Mucus secretion rates for glands stimulated with 3 µM forskolin (Fsk), were WT: 15.5 ± 3.9; 46 glands, n=4, and CF: 5.9 ± 1.8; 27 glands, n=3. To evaluate the consequence of CF mutations on synergistic gland secretion, we combined 100 nM carbachol with 3 µM Fsk. The secretion rates to combined treatment were: WT: 41.9 ± 9.9; 37 glands n=3, and CF: 29.3 ± 10.7; 27 glands, n=3. Unlike tracheal glands, the glands in the turbinates also responded to 100 nM carbachol alone (WT: 271.7 ± 53.5; 9 glands n=1, and CF: 32.3 ± 6.8; 7 glands, n=1). We tested the effects of Substance-P (Subs-P) on nasal glands because it is a potent agonist of airway gland secretion in pigs (Trout et al. Am J Physiol Lung Cell Mol Physiol 2001; 281:639) . Secretion rates to 1 µM Subs-P were small and inconsistent (<12) in both genotypes (46 gland, n=5), unlike the strong responses of piglet tracheal glands (see our companion abstract, Joo et al.). A high concentration of carbachol (1 µM) strongly stimulated gland secretion: WT: 188.9 ± 39.3; 31 glands, n=3, and CF: 64.8 ± 12.5; 26 glands, n=3. As a percent of WT, CF responses to each of the mediators were as follows: Fsk: 38%, Fsk + low Carb: 69%, and high dose carbachol: 34%. The reduced gland secretion in CF piglet turbinate glands to all agonists was apparent even though the CF piglets were older, on average, than the WT piglets (CF: 10.4 ± 7.3 vs. WT: 2.9 ± 1.7 days old). Compared with tracheal/bronchial glands, gland secretion in nasal turbinates from both WT and CF piglets differed in having a lower sensitivity to Subs-P and a higher sensitivity to carbachol. Nasal turbinates have a higher gland density than trachea, while gland sizes appear to be much smaller (~10%), based on secretion rates of maximally stimulated glands. Consistent with tracheal airway glands, nasal glands of CF piglets show mucus hyposecretion. These results from piglet nasal turbinates and tracheal/bronchial airways (Joo et al. companion abstract) suggest that CFTR plays an important role in gland mucus secretion throughout the upper airways.

Supported by CFF and NIH. Introduction: The ability of the airways to move and clear deposited particles is a clear diagnostic indicator of airway health. Prior to the development of synchrotron phase contrast X-ray imaging (PCXI), non-invasive detection and tracking of individual particle movements produced by mucociliary transit (MCT) was not possible. Since MCT is affected in cystic fibrosis (CF), the ability to quantify MCT in CF disease models could assist in examining airway surface dysfunction and responses to treatment. Using PCXI we have dynamically and non-invasively examined the behaviour of individual pollutant particles deposited on the airway surfaces of live intact mice to determine the most suitable particles for transit analysis, and to document particle visibility and behaviour.

Material and Methods: Nasal and tracheal (intubated, flexiVent ventilation) airways of hairless mice were imaged at the SPring-8 synchrotron in Japan. Particles of asbestos, fibreglass, quarry dust and lead (ground galena) as well as reference 14 µm hollow glass beads were suspended in water (nose) or saline (trachea) and instilled into the airways. Images with an effective pixel size of 0.45 µm were captured at regular intervals for up to 40 mins using a high-resolution camera combined with respiratory gating (endinspiratory pause) for the tracheal studies. Motion-detection analysis permitted detection of particle transit behaviour otherwise undetectable in raw images.

Results: In the nose all particles were detectable in vivo, and in the trachea all particles except the asbestos fibers could be detected. In both locations particle transit was heterogeneous: after deposition some particles did not move, while others transited the field of view rapidly. This effect was noted to be dependent on both particle type and size. In both the nose and trachea the particle motion was most apparent early in the imaging period, but in the trachea the majority of deposited particles stopped moving shortly after the imaging began. Most particles did not follow a linear path along the airway: many followed seemingly random, tortuous paths. In the trachea the majority of particles deposited or tracked along the dorsal tracheal wall. Transit of clumped particles was also observed.

Conclusion: PCXI permits detection of particle transit via MCT along live mouse nasal and tracheal airways. The majority of particles deposited in the trachea appeared to immobilise soon after deposition, presumably via trapping on surface mucus/fluid. The tendency for particles to localise along the dorsal wall is consistent with known tracheal MCT patterns. The most visible particles were the hollow glass beads, which could be resolved to less than 10 um. We are continuing with studies to improve our direct and noninvasive MCT assessment methods to assist our understanding and treatment of respiratory diseases such as CF. Chloride secretion by airway epithelial cells is defective in cystic fibrosis (CF). The conventional paradigm is that the CF transmembrane conductance regulator chloride channel (CFTR) is activated through cAMP / protein kinase A (PKA), whereas the Ca 2+ -activated chloride channel (CaCC) is activated by Ca 2+ agonists such as UTP. We report evidence for major crosstalk between Ca 2+ and cAMP signaling in well-differentiated primary cultures of human airway epithelial cells, such that the majority of chloride current elicited by Ca 2+ agonists such as UTP involves Ca 2+ /calmodulininduced activation of adenylyl cyclase I (AC1) with consequent cAMP / PKA signaling. By short-circuit current analysis, Ca 2+ agonists such as UTP, carbachol and ionomycin produced more chloride secretion from CFTR than CaCC, as judged by effects of CFTR and CaCC-selective inhibitors. The CFTR component of chloride secretion was reduced by PKA inhibition, but not by inhibition of PKC or EGFR, and was absent in cell cultures from CF patients. Calcium agonists produced cAMP elevation, which was blocked by inhibition of adenylyl cyclase or calmodulin. We found expression of AC1, a Ca 2+ /calmodulin-stimulated adenylyl cyclase, in human airway epithelial cells and found that selective RNAi knockdown of AC1 abolished almost completely both the cAMP elevation and chloride secretion following Ca 2+ agonists. Immunostaining showed AC1 expression at the apical membrane of cultured airway epithelial cells. Taken together with correlations between cAMP levels and chloride currents, as well as immunoprecipitation studies, we suggest compartmentalized coupling of AC1 and CFTR. Our results provide evidence for major Ca 2+ /cAMP cross-talk in human bronchial epithelial cells involving AC1, and indicate CFTR as the principal chloride secretory pathway in non-CF airways for both cAMP and Ca 2+ agonists.

Background: In cystic fibrosis (CF) lung disease poor hydration of airway surfaces leads to reduced mucociliary clearance, adhesion of mucus and chronic bacterial infection. Levels of inflammatory mediators in sputum from patients with CF might further impair the already defective chloride secretion in the CF airway, promoting further mucociliary dysfunction, infection and inflammation. Inhaled hypertonic saline (HS) appears broadly applicable as an inexpensive therapy for most patients with CF, mildly improving lung function, quality of life and absenteeism and markedly reducing pulmonary exacerbations, without promoting infection or inflammation. Some studies recorded intolerance of regular HS due to sustained cough, presence of new wheeze or rales on the exam and saltiness ranging from 3% to 10% of patients in regular regimen with HS, despite a pretreatment with bronchodilator. Hyaluronic acid (HA) is a naturally occurring polysaccharide, which can be found in numerous tissues and body fluids. HA has several physiological functions and mechanisms, such as a barrier effect and water homeostasis, stabilizing the extracellular matrix, increasing mucociliary clearance and preventing elastin injury.

The aim of our preliminary study is to evaluate the safety and tolerability of 0.1% HA added to HS administered to patients with CF older than 6 years.

Methods: 20 patients with CF regularly treated twice-daily with inhalation of 7% HS were randomized to inhale either 7% HS or HS solution added to 0.1% HA on the following day, in a single blind cross-over design.

The alternate solution was inhaled prior to an identical physiotherapy session and a bronchodilator treatment. The concentration of 0.1% HA is the highest dose permitted for administration to humans in clinical research according to the U.S. FDA and it is the best concentration to be easily and comfortably inhaled. All participants rated the severity of the following symptoms after each dose: cough, irritation (burning in the throat), and saltiness. Each of the symptoms had a 4-point ordinal score ranging from absent to severe. All participants also rated the pleasantness of the treatments using a 5-level, Likert-type scale ranging from very unpleasant to very pleasant.

Results: For each symptom measure, a statistically significant difference was found between patients treated with HS and HA vs HS alone (p< 0.01).

Cough: HS induced cough in 18/20 patients (90%) immediately after therapy with a total score of 32. HS with HA inhalation induced cough in 7/20 patients (35%) after therapy with a score of 8. Throat irritation occurred in 14/20 (70%) after HS with a score of 23. Throat irritation occurred only in 1/20 (5%) after HS with HA (total score 2). Salty taste: All patients treated with HS complained of salty taste while only 2/20 (10%) reported this effect after HS and HA (total score 32). Patients showed pleasantness was significantly higher when treated with HS added to HA.

Conclusion: HA added to HS is generally safe and well tolerated. This novel approach could be a new tool for improving compliance to treatment with HS in collaboration with patients with CF. Clinical trials are needed to support our results.

Crespin, S.; Bacchetta, M.; Scerri, I.; Dudez, T.; Chanson, M.

Functional integrity of the airway epithelium is altered in cystic fibrosis (CF). Epithelial integrity depends on the expression and assembly of specific proteins into specialized junctional structures. Gap junctions, made of connexins (Cx) hexamer, play crucial roles in these interactions by contributing to the ability of cells to share signaling factors directly between adjacent cells. The pattern of Cx expression in Human Airway Epithelial Cell (HAEC) cultures is associated with the differentiation state. Thus, Cx26 is specifically expressed during proliferation phase and its expression decreases to undetectable levels with HAEC differentiation to a polarized airway epithelium. We questioned if a specific pattern of Cx was associated with the airway epithelium repair following injury.

A model of HAEC repair was established by wounding the cultures mechanically. The cultures were followed for several days before fixation, immunostaining and confocal analysis. The Gap Junction Intercellular Communication (GJIC) between adjacent cells was evaluated by microinjection of Lucifer Yellow.

In our model of HAEC repair, Cx26 was transiently re-expressed at the wound area and in basal cells behind the wound. The re-expression of Cx26 was associated with enhanced spreading of the gap junction tracer Lucifer Yellow. In normal HAEC, Cx26 detection was concomitant with the cell ability to migrate and proliferate (as evaluated by Ki-67 detection) to close the gap following injury. The same phenomenon was seen in HAEC from CF patients in higher proportion. Interestingly, the amount of Cx26, the duration of its expression and the number of Ki-67-positive cells were increased, suggesting a hyperproliferative state of the CF airway epithelium. Moreover, the block of Cx26 expression appeared to decrease HAEC proliferation.

These results suggest that gap junctions, and more specifically Cx26, play a role in airway epithelium wound repair and that understanding of the underlying mechanisms may lead to identify new targets for controlling CF HAEC proliferation and differentiation.

Supported by "Vaincre la mucoviscidose" and FNRS. Airway submucosal glands secrete mucus (fluid + macromolecules) in response to various neurotransmitters and mediators such as histamine, which is released mainly by mast cells but also by Pseudomonas aeruginosa (Devalia et al. J Clin Pathol 1989; 42:516) and sometimes by neutrophils (Xu et al. J Exp Med 2006; 203:2907) . Thus, high levels of bacteria and neutrophils in infected CF airways might produce significant levels of histamine in addition to the mast cell source. We studied the effects of histamine on mucus secretion from control and CF human airway glands obtained at lung transplantation. Secretion rates from individual glands were measured using optical imaging (Joo et al. Am J Physiol Lung Cell Mol Physiol 2001; 281:458) . The average responses over a 15-20 min period to 100 µm histamine applied basally were as follows (in nanoliters/gland/min, followed by n of subjects and n glands): control (donor) tracheas: 0.41 ± 0.21, n=9, 67 glands; non-CF disease controls: 0.24 ± 0.07, n=6, 37 glands; and CF: 0.02 ± 0.02, n=2, 18 glands. Each experimental tissue was subsequently tested with 10 µM carbachol and the response to histamine was expressed as a percent of the carbachol response in that tissue. The average histamine/carbachol ratios were: donor control: 25 ± 6%, n=9; non-CF diseases: 15 ± 4 n=6, and CF: 0.004 %, n=2; (CF vs. controls p <0.05); the average carbachol responses of the three groups did not differ. The loss of secretion by CF glands suggests that histamine requires CFTR to produce gland mucus secretion. Histamine elevates intracellular [Ca 2+ ] i in gland cells (Lee et al. J Physiol 2007; 582:1099) , and can potentially activate Ca +2 activated Clchannels, yet the near absence of a response in CF cells suggests that unlike carbachol, but like the [Ca 2+ ] i elevating agent Substance P, it does not do so (Choi et al. J Clin Invest 2009; 119:1189) . As a further test of the CFTR-dependence of histamine-induced gland secretion, we tested a subset of donors to determine the interaction between forskolin (known to stimulate glands in a CFTR-dependent manner (Joo et al. J biol Chem 2002; 277:50710) ) and histamine; we again normalized responses to carbachol responses to control for tissue/gland variability. After a saturating (10 µM) dose of forskolin, histamine failed to produce any increase in secretion rates: forskolin normalized rate was 0.37 ± 0.08; and for 100 µM histamine added on top of forskolin the normalized rate was 0.30 ± 0.08 (n.s.). Lack of additivity between histamine and forskolin further supports a common path through CFTR. Lack of histamine-induced fluid secretion from CF airway glands may contribute to decreased mucus clearance in CF airways. Background: Cystic fibrosis (CF) is characterized by the thick aggregated mucus that affects every mucus-producing organ in the body. Consequently women with CF have a tenacious cervical mucus that retains its viscosity throughout the menstrual cycle. As well as a loss of Clconductance, mutant CFTR channels fail to support bicarbonate (HCO 3 -) transport. We have previously reported that mucus release in the female reproductive tract is critically dependent on HCO 3 -. In the absence of extracellular HCO 3 and in the presence of CFTR inhibitors, mucus release was undetectable [1] . These findings support the recently proposed mucus and HCO 3 hypothesis [2] where HCO 3 may play a critical role in unshielding mucin molecules released into the lumen. In the present study we show cervical mucus release is impaired in ∆F508 mice, whereas fluid secretion is preserved under the same conditions. Methods: ∆F508 mice originated from Case Western Reserve University. The reproductive tracts from wild type (wt) and ∆F508 mice were excised and mounted in a custom-built constant temperature perfusion chamber. Mucus collection: The lumen was perfused with a glucose free Ringer and bathed basolaterally with a HCO 3 or HCO 3 --free Ringer. Mucus content was determined by binding of wheat germ agglutinin lectin to samples. Fluid secretion measurements: One of the uterine horns was cannulated with a fire polished glass capillary. The vaginal end of the tissue was then ligated to form a closed, cannulated sac isolated from the external solution. After equilibrating, fluid secreted into the lumen displaced fluid into the capillary. The rate of fluid secretion was measured under a microscope by monitoring the displacement of the air/liquid meniscus in the capillary.

Results: In contrast to mucus release, Prostaglandin E 2 (PGE 2 ) (1 µM) + carbachol (10 µM) stimulated fluid secretion in wt mice was not dependent on HCO 3 -(79 ± 18 µl. min -1 .g -1 ; HCO 3 -Ringer vs. 70 ± 24 µl. min -1 .g -1 ; HCO 3 free Ringer), or on CFTR (74 ± 15 µl. min -1 .g -1 ; HCO 3 -Ringer + GlyH-101 + Mal H1), but was inhibited by Niflumic acid (0.4 ± 0.1 µl. min -1 .g -1 ; HCO 3 -Ringer). PGE 2 + carbachol stimulated fluid secretion was partially preserved in ∆F508 mice (43.2 ± 19 µl. min -1 .g -1 ; HCO 3 -Ringer). Conversely PGE 2 + carbachol stimulated mucus release was severely impaired in the ∆F508 mice (119 ± 62 % increase; HCO 3 -Ringer) compared to wild type mice (650 ± 77 % increase; HCO 3 -Ringer). Conclusions: Defective cAMP-mediated fluid secretion is a characteristic of all exocrine systems affected in CF, suggesting abnormal mucus in CF may be simply due to impaired fluid secretion. We show that PGE 2 + carbachol stimulated mucus release, which is CFTR and HCO 3 dependent, is impaired in the ∆F508 reproductive tract, whereas fluid secretion is partially preserved under similar conditions. Our findings suggest that rather than a simple loss of fluid secretion, the loss of CFTR dependent HCO 3 secretion is an important factor contributing to depressed mucus release. The bronchial airways of the lung have the potential to both secrete and absorb liquid. Liquid absorption occurs across the surface epithelium of the bronchi, whereas liquid secretion occurs from both the submucosal glands and the surface epithelium. Previous studies documented that the net liquid secretion induced by glandular secretagogues can be measured in cannulated bronchial airways excised from the lungs of domestic pigs. This technique has now been modified to permit measurements of both unidirectional liquid secretion and unidirectional liquid absorption across porcine bronchi. From each pig, three bronchial airways were dissected from the lungs, cannulated, and immersed in bicarbonate-buffered Krebs solution. Two airways were filled with a known volume of Krebs solution. One of these liquid-filled airways was used to measure net liquid transport. The other liquid-filled airway was pretreated with 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB) to block unidirectional anion and liquid secretion to obtain a measurement of unidirectional liquid absorption. Unidirectional secretion was calculated as the difference between the net change and the unidirectional absorption. A third airway remained air-filled for comparison. The airways were incubated for 2 h after which the rates of liquid secretion or absorption were calculated based on changes in luminal liquid volumes. These liquid transport responses were assessed under the control condition (no secretagogue) or when airways were treated with 10µM acetylcholine (ACh), 1µM substance P, or 100nM vasoactive intestinal peptide (VIP). Under the control condition, the bronchi exhibited unidirectional absorption of liquid (-4.25±0 .62 µl᭹cm -2 ᭹h -1 ) while unidirectional secretion was essentially absent. ACh, substance P, and VIP induced unidirectional secretion of +14.55±1.37 µl᭹cm -2 ᭹h -1 , +12.18±1.36 µl᭹cm -2 ᭹h -1 , and +4.42±1.26 µl᭹cm -2 ᭹h -1 , respectively, whereas unidirectional liquid absorption remained near control levels (-5.74±0 .89 µl᭹cm -2 ᭹h -1 , -4.37±0.49 µl᭹cm -2 ᭹h -1 , and -4.72±0.81 µl᭹cm -2 ᭹h -1 , respectively). Net liquid flux in the presence of the secretagogues was slightly higher (in the direction of secretion) in the air-filled bronchi than in the liquid-filled bronchi. These data demonstrate that both net and unidirectional liquid fluxes can be accurately measured in intact bronchial airways from pigs.

Supported by NIH HL063302. Adenosine and ATP acting on airway epithelial cell surface purinergic receptors regulate key components of mucociliary clearance (MCC), i.e., ion transport, mucin secretion, and ciliary beat frequency. Inefficient MCC results in mucus adhesion, inflammation, and infection contributing to the pathogenesis of cystic fibrosis. While remarkable progress has been made in understanding how these epithelial functions are regulated by purinergic receptors, it remains poorly understood how ATP reaches the airway surface to accomplish extracellular signaling.

Recently, we discovered that the protease thrombin promotes robust RhoA-dependent release of ATP from primary cultures of well-differentiated human bronchial epithelial (HBE) cells. Thrombin-promoted ATP release was associated with connexin/pannexin hemichannel opening (Seminario-Vidal et al. J Biol Chem 2009 May12 [Epub ahead of print] ). In the present study, we asked whether RhoA and connexin/pannexin hemichannels are involved in mechanically-induced ATP release from HBE cells. Hypotonic challenge promoted robust ATP release that was accompanied by activation of RhoA and enhanced phosphorylation of myosin light chain (MLC), a downstream effector of Rho kinase (ROCK) and MLC kinase (MLCK). Inhibition of ROCK and MLCK diminished hypotonicity-induced ATP release in primary HBE cells. Furthermore, hypotonicity-induced ATP release was drastically impaired in lung epithelial cells transfected with RhoAT19N (RhoA dominant negative mutant). These results suggest that activation of RhoA/ROCK and MLCK are necessary for ATP release in hypotonicity-stimulated HBE cells. Hypotonicity-stimulated ATP release was also decreased by non-selective connexin/pannexin hemichannel inhibitors. Confocal microscopy studies illustrated that hypotonic challenge elicited a rapid uptake of the hemichannel-permeable reporter dye propidium iodide. Hypotonicity-induced propidium iodide uptake kinetics was similar to that of ATP release and, like ATP release, it was inhibited by hemichannel blockers, ROCK inhibitors, and RhoAT19N. RT-PCR analysis of HBE cells revealed the expression of Panx1, Cx26, Cx30.3, Cx31, Cx31.1, and Cx43, but not Panx2, Panx3 or Cx32 in these cells. An initial screening indicated that Cx43 blocking peptide (i.e., Gap26) or 10Panx1blocking peptide (but not with their scrambled versions) reduced hypotonicity-induced ATP release from primary HBE cells. Taken together, our data suggest that connexin/pannexin hemichannels act as ATP conduits in hypotonicity-stimulated airway epithelial cells and that hemichannel opening is regulated by MLCK-and RhoA/ROCK. Supported by CFF SEMINA08F0 and NIH P01HL34322. Proper regulation of the airway surface liquid (ASL) volume is critical to allow for effective mucus clearance from the lung. The ASL, which is composed of periciliary liquid and mucus, is regulated by the active transport of sodium and chloride which establishes osmotic gradients that drive fluid flux across the airway epithelium. Despite the importance of the ASL to mucus clearance, few labs have the resources to study the volume or height of this liquid layer. We observed that when primary human airway cells are cultured on an air-liquid interface (ALI), a liquid meniscus typically develops at the edge of the culture insert. We reasoned that the size of the fluid meniscus is likely determined by the ASL volume and could be measured as an index of the epithelial surface hydration status. Therefore, we developed a simple method to measure the meniscus length and determined whether this measurement changes with perturbation to the ASL volume.

Primary cells from >5 CF and non-CF lung donors were cultured on an ALI for 1 month to allow for the development of a mucociliary phenotype. To measure the meniscus length, the cultures were viewed on inverted microscope with a 4X objective and images were acquired from the perimeter of the culture insert. The width of the meniscus was measured using a custom automated script in ImageJ.

In the first series of experiments we determined whether the meniscus length measurement could be used to differentiate between CF and non-CF cultures. Under ALI conditions, the meniscus length was 501.0 ± 10.5 in non-CF cultures vs 216.8 ± 5.8 µm in CF cultures (p < 0.001, n> 20 cultures derived from 5 donors). Therefore, the meniscus length measurement is sensitive enough to differentiate CF from non-CF cultures. Next, a series of experiments was performed to determine whether the meniscus length could detect differences in ASL volume following volume expansion with isotonic (KBR) or hypertonic (KBR + 300 mM mannitol) fluids. A 2.5 µl bolus of the solutions was applied to the apical surface and changes in the meniscus length were measured. Two hours following the fluid bolus, the meniscus length increased 1.8 ± 0.1 fold under isotonic conditions compared to a 2.6 ± 0.2 fold increase with the hypertonic fluid (p= 0.002, n=12 cultures from 3 non-CF tissue donors). These findings support our hypothesis that the meniscus length conveys information regarding the ASL volume.

In summary, we have developed a novel method that can be performed easily with readily available resources to assess the ASL volume. These tools can be used to rapidly assess new therapeutic agents developed to restore ASL volume to the CF airway. We are currently investigating whether mathematical models can be developed to determine the volume and surface tension properties of the ASL based on the shape of the surface meniscus.

Supported by the NIH, CFF, and ALA.

the crucial role of ASL in maintaining pulmonary health, little is known about the mechanisms of electrolyte fluid transport in the native small airways. It is believed that the ASL volume in the small airways primarily reflects a balance between the secretory and absorptive functions. However it is still very unclear as to whether the absorptive and secretory functions are mediated by the same or by distinctly different cell types. This study is designed to test the hypothesis that distinctly different cell types mediate the absorptive and secretory functions in small airways. Using intracellular microelectrodes we have characterized the electrophysiological properties of different airway epithelial cells. Methods: Segments of the small airways (~1mm in diameter) from pig lungs were cut open to expose the luminal surface for random impalements by microelectrodes (Reddy and Quinton, 1992) . The epithelium was continuously perfused with standard phosphate buffered Ringer's solution maintained at 37 o C. We investigated the effects of epithelial sodium channel (ENaC) blocker amiloride (10 -5 M) and the CFTR agonists IBMX+ Forskolin (10 -6 M each) on the intracellular potentials (Vm) of airway epithelial cells.

Results & Discussion: Small airways appear to be composed of cells with different physiological properties. Approximately 41% of cells responded to amiloride by hyperpolarizing Vm (DVm= -6.4±0.5 mV, n=number of cells = 34). Approximately 55% of cells responded only to cAMP agonists (IBMX+Forskolin) by depolarizing cell potentials (DVm= 5.6±1.1, n=31 cells). Approximately 10% of the cells responded to both amiloride and cAMP agonists and ~11% of the cells did not respond to either amiloride or cAMP agonists. Results from immunolabeling of Na/K/2Cl co-transporter appear to indicate that this transporter (which plays a characteristic role in secretory cells) is predominantly confined to a specific group of cells located in the pleats but not in the ridges of the epithelial folds of small airways. Assuming that ENaC is critical for Na + transport and hence electrolyte fluid absorption, these results seem to indicate that only about half of the bronchiolar epithelial cells are engaged in fluid absorptive function. Furthermore, since about half of the cells responded to cAMP agonists but not amiloride seems to indicate that these cells may be involved in a function other than absorption. Further studies are required to determine whether these amiloride insensitive cells in fact express Na/K/2Cl co-transporter and therefore are responsible for secretory function.

Conclusions: Small airways appear to contain separate populations of secretory and absorptive cells that maintain appropriate airway hydration by balanced absorptive and secretory activities.

Acknowledgements: The authors are grateful to Mr. Kirk Taylor for expert technical assistance.

Funded by NIH-RO1 DE14352, and the Nancy Olmsted Trust and the Cystic Fibrosis Foundation. Most strains of P. aeruginosa secrete the virulence factor pyocyanin (Nmethyl-1-hydroxyphenazine; mw 210) that provides growth and survival advantages for P. aeruginosa in the lungs of people with cystic fibrosis (CF). Pyocyanin is cytotoxic and concentrations in pulmonary secretions from CF patients have been shown to reach levels up to 100 µM. Pyocyanin has the ability to penetrate cell membranes, and is known to generate reactive oxygen species (ROS) by its ability to redox cycle with cellular electron donors, such as NADPH and glutathione. Since CFTR couples ATP hydrolysis to the extracellular transport of chloride and alters its anion selectivity for bicarbonate in response to metabolic changes, we investigated the central role of pyocyanin on CFTR Cltransport function (Ussing chamber assay) in CFTR-corrected CF bronchial epithelial cell monolayers (wtCFTR-CFBE41o-) and delineated its impact on key amino acid metabolites involved in the transmethylation, transsulfuration, and glutathione synthesis pathways using a high-throughput liquid chromatography linked tandem mass spectroscopy technique (LC/MS/MS). Acute exposure of the apical membrane to pyocyanin (100 µM) for 2-3 hours caused a gradual decline of CFTR Clcurrents that were stimulated by forskolin (to 86%), genistein (to 76%) or L-ascorbate (to 90%). Interestingly, CFTR Clcurrents stimulated by L-ascorbate were more rapidly blocked by pyocyanin and the faster rundown was likely due to the fact that L-ascorbate enhanced the production of ROS by serving as an electron donor for pyocyanin. These data suggested that prolonged exposure to pyocyanin induced increased ROS production which affected normal CFTR function.

To measure the metabolic impact of pyocyanin, wtCFTR-CFBE41ocells were exposed to pyocyanin for 24 hours. Intracellular ATP levels and steady-state concentrations of the major aminothiols glutathione (GSH) and cysteine (Cys) showed a marked decline, whereas the oxidized form glutathione disulfide (GSSG) and cystine (Cyss) increased following pyocyanin treatment thereby shifting the cysteine and glutathione redox states to a more oxidized state. Metabolites of the GSH synthesis pathway were characterized by a marked decline of glutamate. Metabolites of the arginine pathway were characterized by an increase in arginine, spermidine, and proline. Metabolites of the transmethylation pathway were characterized by an increase in S-adenosylmethionine (SAM), the major biological methyl donor.

In summary, these data show that pyocyanin leads to alterations of both CFTR function and the metabolic state of CFTR-corrected CF bronchial epithelial cells and suggest that metabolic changes might occur in Pseudomonas-infected CF airways that hamper the treatment of the underlying chloride channel defect by CFTR activators. Supported by NIH grant P01 AT002620, PACFI, ENF, and CFF. Nasal polyposis (NP) is a useful model to study pathogenesis of chronic airway inflammation. NP could be either of unknown origin (primary NP) or secondary to cystic fibrosis (CF). To investigate the pathogenesis of human nasal epithelial cells (HNEC) from NP, we first performed relative protein quantitation by iTRAQ labeling followed by mass spectrometry (MS/MS) analysis. Among almost 600 proteins found to be differentially expressed, 158 were quantified with at least 2 peptides. The expression of 8 proteins was found to be significantly up-regulated in cells derived from primary and CF NP compared to control (mucosa without inflammation). One of the up-regulated proteins was GRP78, a protein involved in unfolded protein response (UPR). This result prompted us to further characterize UPR in primary and CF NP. Immunoblots were used to evaluate GRP78 expression at 7 and 21 days. We found an over-expression (3-4 fold) of GRP78 in primary and CF NP, compared to controls. To confirm a possible UPR induction in primary and CF NP, we performed immunoblots with three other UPR markers (calreticulin, GRP 94 and SERCa2). These 3 markers were found to be up-regulated in primary and CF NP (3-4 fold) at 7 days of culture, as compared to controls. Searching for the origin of UPR we tested a possible involvement of constitutive abnormally folded proteins. We performed the same experiments after 21 days of culture and found that the expression of 4 markers of UPR was unchanged in primary and CF NP vs. controls, showing a normalisation of UPR state with time of culture and thus suggesting another UPR origin. Knowing that aggregation of glycoproteins may trigger UPR, we tested whether MUC5AC, a major mucin expressed in airway cells, aggregates in endoplasmic reticulum after 7 days of culture. MUC5AC expression was similar after 7 and 21 days of culture, suggesting that this mucin is not involved in the trigger of UPR. To investigate a possible NP-dependent UPR induction, we performed immunoblots of GRP78 on proteins from primary culture of HNEC obtained by nasal brushing in patients with chronic rhinosinusitis or primary NP and in controls. GRP78 was found to be over-expressed in primary NP compared to control patients (3-4 fold) but was unchanged in chronic rhinosinusitis vs. controls, thereby showing that inflammation alone is not responsible for UPR and suggesting the involvement of UPR induction in the pathophysiology of NP. To investigate the connection between UPR response and inflammation, we measured secreted IL-8 by ELISA from primary cultures of primary NP and control patients at 7 and 21 days of culture. IL-8 levels were found to be increased after 7 days in primary NP culture vs. control (4-5 fold) , and normalised after 21 days of culture. These data indicate that inflammation may be in part consecutive to UPR response. Altogether, our results strongly suggest that UPR is related to inflammatory microenvironment and plays a role in the pathogenesis of primary and CF NP. The mechanism that triggers UPR in NP remains to be discovered.

Supported by ANR 05MRAR022.

In cystic fibrosis (CF) the CF transmembrane conductance regulator (CFTR) basic defect is currently targeted by multiple approaches to restore CFTR chloride (Cl -) channel activity using correcting and potentiating pharmacotherapeutic compounds. These drugs are evaluated and pharmacologically optimised in vitro in studies on cell lines, tissue cultures and animal models. However, CFTR modulators might have a different efficacy in native human CF epithelia. Availability of native human CF tissues is relatively limited, and size and condition are important quality markers. Hence, the aim of this study was to establish lung explant epithelia from CF and disease control patients receiving lung transplantation as a new ex vivo model for preclinical priorisation of CFTR pharmacotherapy.

We included native human bronchial tissue sheets from 57 lung explants (CF, COPD, pulmonary fibrosis) and donor bronchi obtained during lung transplantation at the Hannover transplant center in this study so far. Multiple (n=8) tissue pieces per subject were prepared immediately after surgery, transepithelial short-circuit current (I sc ) in perfused Mini Ussing chambers was registrated and structure control was obtained by histopathology.

The performed preparation technique resulted in tissue segments with intact respiratory epithelia including submucosal glands. Basal tissue bioelectric properties (transepithelial resistance 15.2 ± 5.4 Ωxcm 2 ; mean ± SD), I sc responses to inhibition of the epithelial sodium channel ENaC and stimulation of CFTR and alternative Clsecretion (CaCC) in CF and disease control bronchi showed reproducible results in a total of 456 bronchial tissue segments (∆I sc amiloride -3.1 ± 1.7 vs. -1.5 ± 2.6 µA/cm 2 ; ∆I sc forskolin/carbachol -11.2 ± 15.3 vs. 4.5 ± 4.9 µA/cm 2 ; ∆I sc ATP -2.4 ± 1.6 vs. -3.4 ± 3.1 µA/cm 2 ).

The results of this pilot study confirm the feasibility of ex vivo Clsecretion measurements in native human lung explant tissue as an additional resource of human CFTR and CaCC relevant tissue. Data about variability, long-term ex vivo vitality and possible limitations of the method will be added, aiming to investigate the effects of most promising CFTR correctors/potentiators and CaCC activators. Our strategy can be an essential step in the translation of small molecule compounds into future clinical trials aiming to rescue the CF basic defect.

Supported by a financial grant from Mukoviszidose e.V. (the German Cystic Fibrosis Association) and an EuroCareCF training grant. the cholesterol signaling pathway contribute to inflammation. Focusing on modulation of plasma membrane cholesterol content in vivo provides an accessible biomarker for cholesterol processing in CF cells and tissues.

Resveratrol is a polyphenol able to regulate cholesterol processing in cells, however, in vivo data is lacking. Our laboratory has previously shown that resveratrol decreases cholesterol synthesis and plasma membrane cholesterol content in vitro, but studies had not been performed in vivo. The primary aim of this study is to investigate plasma membrane cholesterol content in a CF mouse model. A second aim is to identify the mechanism by which resveratrol exerts its effect on cholesterol homeostasis. We hypothesize that resveratrol will decrease plasma membrane cholesterol content in vivo and that resveratrol modulates cholesterol content through the AMPactivated protein kinase (AMPK) pathway.

Mice lacking CFTR expression (CFTR tm1Unc ) and CFTR WT mice were treated with resveratrol (5% solution in drinking water) or vehicle. Nasal epithelium was resected for evaluation of plasma membrane cholesterol content, measured by direct electrochemical evaluation. Nasal epithelial cell membrane cholesterol content is significantly decreased in resveratrol-treated CF mice compared to vehicle-treated mice. A 1.8-fold decrease in membrane cholesterol content is seen in resveratrol-treated CF mice. Similar results are seen in WT resveratrol-treated mice.

Resveratrol treatment is known to result in activation of AMPK, a protein known to decrease cholesterol synthesis. The possibility that resveratrol was impacting cholesterol regulation in CF cells through AMPK activation was examined by monitoring membrane cholesterol content. Human epithelial cells were treated with resveratrol (50 µm for 24 hours). In the presence of AMPK inhibition (AMPKi compound C, 10 µm for 24 hours), the resveratrol-mediated drop in membrane cholesterol is blocked. To verify that resveratrol is able to activate AMPK in these cells, phosphorylated AMPK protein content was measured by Western immunoblot and normalized to total AMPK content. Resveratrol-treated cells show a higher phosphorylated AMPK to AMPKα ratio compared to untreated cells (1.8-fold +/-0.5 over untreated cells). A similar increase in AMPK phosphorylation is seen in resected lungs from resveratrol treated CFTR deficient mice.

In summary, plasma membrane cholesterol content is modulated by resveratrol in vivo in a CF mouse model, which supports previous data in CF cell models. This provides a measurable biomarker for cholesterol processing in CF. Because resveratrol impacts cholesterol processing and is a known activator of AMPK, it is a unique compound for investigation of cholesterol content in CF. These data demonstrate that resveratrol leads to AMPK phosphorylation and that the ability of resveratrol to regulate cholesterol homeostasis is likely through AMPK, although the exact mechanisms of AMPK modulation of these pathways will require further study.

Supported by a grant from the CFF. The better understanding of pathophysiological mechanisms involved in the first stages of CF lung disease development requires the analysis of respiratory epithelium samples obtained from young asymptomatic CF infants. Since nasal epithelium is generally considered to be representative of the lower airway epithelium, the present work aimed to harvest airway cells by nasal brushing. These native cells were then immediately used for the study of the airway epithelium functions. Airway epithelial sheets collected from brushings were analyzed by videomicroscopy for ciliary beating frequency measurement, for assessment of the cAMP dependent chloride efflux and for measurement of the gap junctions functionality by using a fluorescence recovery after photobleaching (FRAP) technique. In 6 CF infants, nasal brushing was performed within 2 months after birth and then when children were 1 year old. Six non-CF infants were also studied as control. At the time of nasal brushing, all non-CF patients were free of clinical signs of respiratory infection, and CF patients had no respiratory symptoms. We observed that the ciliary beating frequency was similar in the CF infants (12.5±1.1 Hz) and in the non-CF infants (12.7±0.5 Hz). The chloride efflux in the CF infants was 3.5 fold decreased (p=0.01) compared to the non-CF infants. This decreased chloride efflux was also observed in the airway cells obtained during the second brushing of CF infants. Large inter-individual variations of chloride efflux were noticed in the CF group (coefficient of variation: 94%) compared with the non-CF group (coefficient of variation: 34%). The gap junction functionality in airway epithelial cells from CF infants was not significantly different compared with the non-CF infants.

The present data demonstrate that, prior to any lung disease development in CF infants, apart from the CFTR defect, the airway epithelium integrity seems to be preserved.

Supported by Association Vaincre La Mucoviscidose. Robert A Good, MD was the first to know that the mucus was dehydrated and so prevented the bacteria reaching the immune system so that the mucus in the airways was preventing the dead bacteria from reaching the lymph nodes and therefore not producing normal amounts of gamma globulin. Good assigned WJ Warwick to find a treatment for this dry mucus problem.

When LeRoy Matthews learned of Good's idea he had all his CF Clinic's CF patients sleep all night in a Mist Tent filled with a fog of saline. His patients' survival increased from 2 years to 10 years. When his claim was real most CF Centers adopted the mist tent treatment and confirmed that it improved survival. Mist tents were banned when the majority of these patients by age 12 years developed Pseudomonas infection in their lungs and those infections were blamed on the Mist Tent.

Next hypertonic saline aerosol was shown to increase the water in the airways and that regular use improved CF patients' health and survival.

Genetic research of the CF Foundation has led to the discovery of the CF gene which facilitates the cells' secretion of water. When that gene is added to the CF mucus cell these cells will produce normal amounts of water in the mucus. This normal mucus secretion lasts until the treated cells die. This work could be improved by repairing the abnormal gene instead of adding the extra normal gene.

In Minnesota, our Defense of the Lungs Project is now on a study of how the high frequency chest compression (HFCC) trapezoid waveform can change the exocrine mucus secreting cells to secrete more water, will this make normal mucus secretion, and how long this normal secretion continues after the patient switches to a HFCC sine wave form system.

In one early study, we measured the water content of sputum when the patients were using a sine waveform system and again after these patients' sputum was collected after the patients used a trapezoid HFCC system for 30 minutes twice a day for 4 or 5 days.

In another early study we found that the increased water in their sputum lasted for at least 5 days after they stopped using the trapezoid system therapies and switched to manual therapy.

These patients' sputum showed more increased water in their sputum after 4 to 5 days of twice a day 30 minute treatment with a trapezoid wave form machine and that the increased duration can last at least of 5 days switching to manual therapy.

Our research is to find (a) How many days of treatment are needed to increase the sputum's water content, (b) What the stability, the mean and SD will the increased water content show over a long period, and (c) What changes and the time when these changes occur as CF patients switch back to the sine waveform to find how many days for the water content to reach a new mean and SD.

Xu, X. 2 ; Jackson, P.L. 2 Chronic neutrophilic inflammation is a common feature of a variety of lung diseases (such as chronic bronchitis and cystic fibrosis (CF)), leading to aggressive airway remodeling and increased morbidity/mortality. Recently our group has characterized N-acetyl Pro-Gly-Pro (Ac-PGP) as an important neutrophil chemoattractant in CF lung disease. Ac-PGP is generated from the breakdown of extracellular matrix (ECM) by the actions of matrix metalloproteases (MMPs). MMP-9 is present in the tertiary granules of neutrophils and can be released following stimulation by inflammatory mediators. In the present report, we explore the ability of Ac-PGP to cause granule release from PMNs and determine if Ac-PGP-induced granule release could result in a feed-forward cycle by releasing the enzymes (i.e. MMP-9) required for its production. Freshly isolated human neutrophils were stimulated for 0, 15, 30 and 45 min with 1.0 mg/ml Ac-PGP. The cell-free supernatants were then collected and assayed for the presence of MMP-9 by gelatin zymography and ELISA. The results showed that Ac-PGP induced significant release of MMP-9 as early as 15 min post-treatment with the maximal MMP-9 release at 30-45 min. This release was recapitulated from quiescent PMNs isolated from mouse bone marrow. Both the MAPK/ERK pathway and p38 pathways are activated with Ac-PGP stimulation of PMNs. However, significant inhibition of MMP-9 release is observed only with ERK-pathway blockade. These data indicate that ECM-derived Ac-PGP could result in a feed-forward cycle by releasing MMP-9 from PMNs, augmenting PGP production and neutrophil influx. In addition, these results also highlight an important intracellular signaling pathway of MMP-9 release from PMNs. Targeting this feed-forward pathway in CF lung disease may serve as one mechanism to modulate disease progression. Kreda, S.M.; Seminario-Vidal, L.; van Heusden, C.; O'Neal, W.K.; Jones, L.C.; Boucher, R.C.; Lazarowski, E.R. Cystic Fibrosis/Pulmonary Research and Treatment Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA Purinergic regulation of native defence activities at the airway lumen is achieved by the release of ATP and other nucleotides from the underlying epithelial cells. However, there is no consensus about the mechanism(s) of nucleotide release. We have illustrated that ATP is secreted from goblet airway epithelial cells during Ca 2+ -regulated mucin exocytosis, suggesting that ATP is a releasable cargo-molecule within the mucin granule (Kreda et al. J Physiol 2007; 584:245) . This hypothesis was reinforced by illustrating that isolated mucin granules contain ATP and other adenyl-nucleotides (Kreda et al. Pediatr Pulmonol 2008; S31:246) . Thus, ATP released from mucin granules likely signals through purinergic receptors on neighbouring (ciliated) cells to promote ion/water transport for hydration of secreting mucins in the airway lumen. Here, we describe a receptor-mediated mechanism that promotes coordinated ATP and mucin release during inflammation in obstructive lung disorders. Thrombin and neutrophil elastase are human proteases abundant in lumenal secretions of airways undergoing inflammation, and are agonists for the protease activating receptors (PAR) 1 and 2, respectively. RT-PCR analysis revealed the expression of PAR-1 and PAR-2 in Calu-3 and primary human bronchial epithelial (HBE) cells. Thrombin and elastase applied basolaterally stimulated the simultaneous and lumenal release of mucins (assessed by immuno-slot blot) and ATP (quantified by the luciferase assay) from Calu-3 and HBE cells. PAR-1 and PAR-2 selective peptide agonists mimicked the proteases' response, suggesting that PAR-1 and PAR-2 activation mediated the ATP and mucin release stimulated by thrombin and elastase, respectively. Thrombin, elastase, and PAR-1 and PAR-2 activating peptides elicited mucin granule exocytosis as determined by confocal microscopy studies of cells labelled with FM 1-43 and MUC5AC antibodies. PAR-mediated mucin and ATP release was dependent on intracellular Ca 2+ mobilization and actin cytoskeleton, since BAPTA-AM and cytochalasin D blocked these responses. We recently established that Rho kinase and myosin light chain (MLC) kinase regulate thrombinpromoted ATP release in airway cells (Seminario-Vidal et al. J Biol Chem 2009 May12 [Epub ahead print] ). Inhibitors of Rho and MLC kinases blocked not only ATP release, but also mucin granule secretion in PAR-stimulated Calu-3 and HBE cells.

In sum, airway goblet cell mucin granules may be an important source of ATP release in airways undergoing inflammation and goblet cell hyperplasia. Co-release of ATP and mucins occurs upon activation of PAR-1 and PAR-2 in a Ca 2+ -, Rho and MLC kinase-, and actin cytoskeleton-dependent manner. The potential pathophysiological consequences of protease-induced ATP release with mucin secretion in airways undergoing inflammation remains to be elucidated.

Supported by NIHP01-HL34322. Mucin proteins are the main macromolecular component of airway mucus, a fundamental constituent of the mucosal innate defense system, whose failure is responsible for the development of severe pulmonary disease in cystic fibrosis (CF). Our current hypothesis is that loose physical interactions between secreted mucins (Muc5AC and Muc5B) and membrane-tethered mucins (Muc1, Muc4, Muc16), in the context of normal airway surface liquid, allow for efficient airway mucus clearance via the action of beating cilia and cough. In the context of a dehydrated airway surface, such as would be expected in CF lungs, secreted mucins become adherent to the underlying airway epithelium, possibly due to inappropriate interactions of secreted mucus with membrane-tethered mucins, which are predicted to form a significant component of the epithelial glycocalyx via their large extracellular glycosylated domains. To begin to unravel the significance of the interactions between secreted and membrane-tethered mucins, we have utilized mice deficient in Muc1 and Muc4 in the context of the Scnn1btransgenic mouse model, in which airway-targeted overexpression of the βsubunit of the epithelial sodium channel (βENaC) results in airway surface dehydration, early postnatal death due to airway mucus obstruction, and chronic neutrophilic inflammation. We hypothesized that, in the absence of Muc1 and Muc4, secreted mucus would be less adhesive and better cleared from the airways, resulting in less inflammation and amelioration of the chronic lung disease that develops in this model. Contrary to our predictions, we detected no significant difference in the phenotype of the Scnn1b-Tg mice in the absence of either Muc1, Muc4, or in the absence of both Muc1 and Muc4. Survival did not significantly improve, nor did any of the parameters that are normally measured to monitor disease phenotype, including presence of airway inflammation and infiltrating inflammatory cell profile, occurence of mucus plugs, and incidence of air space enlargement. We have previously shown that deletion of Muc4 can lead to transcriptional up-regulation of Muc16, and so we suspect that functional redundancy is responsible for the lack of phenotype in Muc1/Muc4 Scnn1b-Tg mice. To address this issue, we have begun breeding the Scnn1b-Tg mice to the Muc16 knockout mice. Preliminary results of this cross are pending. We anticipate that the availability of all of these models, in conjunction with mouse models deficient for Muc5AC and Muc5B, will allow us to study the role of individual mucins in the adhesive properties of airway mucus under normal conditions and in the presence of CF-like airway surface dehydraton.

Funded by CFF grant to W.K. O'Neal (CFF R026-CR037 Project 2). The gel-forming mucins are very high molecular weight (6-100 MDa), heavily glycosylated proteins which are stored in mucin secretory granules (MSGs) and secreted by goblet cells from the airway surface epithelium or from mucous cells in submucosal glands. The principal airway mucins, MUC5AC and MUC5B, are the precursors to normal airway mucus, which is a key part of the mucociliary clearance system, the first line of innate defense in the airways. The molecular mechanisms by which these mucins are packaged in MSGs and released by exocytosis, and how the released mucins mature to form a functioning mucus layer are not understood. We have therefore worked to develop a method to isolate intact MSGs, to be used as a common technical approach to all of the above questions. Starting with extracts of human bronchial epithelial (HBE) cells, or of N3T cells, a cell line derived from HBE cells which exhibits a goblet cell phenotype, equilibrium centrifugation in a continuous Iodixanol (OptiPrep®) gradient (5-33%) , was used to separate MSGs from mucins (released or secreted) and other cell organelles. The results revealed that the buoyant density of MSGs is only slightly higher than that of secreted mucin, ~1.1 and ~1.05 g/ml, respectively (note: the usual buoyant density reported for mucins isolated by equilibrium centrifugation in CsCl 2 is ~1.4 g/ml). To achieve a faster and higher degree of separation, rate zonal centrifugation was used to take advantage of the differences in sedimentation rates between MSGs and secreted mucins, rates that are heavily influenced by size and shape. Antibodies specific to MUC5AC or MUC5B were used to probe fractions bound to nitrocellulose in a slot blot apparatus. The MSG-rich fraction identified was also positive for Rab3D, a well-known marker for mucin and many other secretory granules. Light microscopy of the MSG-rich fraction revealed µm-sized spherical organelles which trapped acridine orange, indicating the low pH environment expected for MSGs. By electron microscopy, a dense, fibrous material was clearly associated with the organelles when they were disrupted by low concentrations of detergent. This MSG purification method represents a tool that can be used to study many aspects of mucin biology, including how it is unpackaged following release from the granule. This study was supported by a grant from the Mucus Consortium funded by Cystic Fibrosis Foundation Therapeutics. Defective electrolyte transport by airway epithelia may play an important role in the pathogenesis of cystic fibrosis (CF) lung disease. To better understand how loss of CFTR affects transepithelial ion transport, we investigated airway epithelia in a porcine model of CF. We studied epithelia from pigs within ~12 hr of birth and assessed electrolyte transport in vivo in the nose and trachea and in vitro with freshly excised tissue and epithelial cultures from the nose, trachea and small bronchi. There were several general findings. a) We observed no consistent differences between CFTR +/+ and CFTR +/epithelia, suggesting that reduced amounts of CFTR are sufficient for normal transport. b) After inhibiting short-circuit current (I sc ) with amiloride, increasing cellular cAMP by adding forskolin and IBMX stimulated I sc in wild-type and CFTR +/epithelia, but generated little or no current in CFTR -/epithelia. In vivo studies of V t yielded similar results. In addition, HCO 3 transport was eliminated in CFTR -/epithelia. These data indicate the loss of CFTR function in CFTR-null airways. c) We observed transient electrical responses to apical addition of ATP that did not differ by genotype. d) Under basal conditions, V t was greater in vivo than in freshly excised tissue. Our studies suggest a significant component of edge damage that increased transepithelial electrical conductance (G t ) in freshly excised tissue. The G t of cultured epithelia was much lower than that of freshly excised epithelia. e) Electrolyte transport by trachea and lower airways was similar, but there were differences between those epithelia and nasal epithelia. Compared to epithelia from intrapulmonary airways, cultured nasal epithelia had greater values of basal I sc and G t ; this was the case in both CFTR -/and control epithelia. In wild-type epithelia, nasal cultures also showed a smaller current increase with elevations of cAMP than did epithelia from intrapulmonary airways. These data suggest physiological regional differences, even in newborn pigs. In summary, these studies allow a comparison between in vivo and in vitro epithelia and between different airway regions that has not been possible in humans. They also provide an opportunity to investigate transport at a time when epithelia do not manifest changes secondary to disease. Similarities between the function of human and porcine airway epithelia suggest porcine airway epithelia should be of value for better understanding the early development of CF lung disease. Hill, Chapel Hill, NC, USA; 2. Department of Computer Science, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA The critical dysfunction in cystic fibrosis (CF) is an increase in mucus viscosity, which disrupts healthy mucocilliary clearance in the airway. For this reason, measurement of mucus viscosity is key to quantifying CF pathology and developing new CF therapies. Unfortunately, mucus specimens are difficult to obtain, viscometry is low-throughput, and volume sizes are small.

We demonstrate a high throughput rheometer that uses small specimen volumes (< 25 µL). It conforms to commercial high throughput screening standards and is scalable to 96 wells. This Multiforce High Throughput System (MHTS) uses magnetic fields to non-invasively apply forces to magnetic particles of sizes ~1-10 µm. The MHTS can be used to make mechanical measurements of cells, tissues, and polymers, and also to study the transport of magnetic probes through these materials. For the characterization of mucus, we show that driven bead rheology can measure the linear viscoelasticity, shear thinning and microheterogeneity. Background: SLC26A9 is a member of the SLC26 family of anion transporters. SLC26A9 has been reported to function as a chloride-bicarbonate exchanger, a chloride channel, and a sodium-anion cotransporter. SLC26A9 has been found in trachea, bronchi, and small airway surface epithelium. A549 cell line is a human-derived lung adenocarcinoma cell line that does not express CFTR.

Methods: Cells were assayed for chloride/bicarbonate exchange activity by loading them with the pH-sensitive dye, BCECF, and measuring ratiometric flouresence in the presence and absence of chloride in the bathing solution. A549 cells were transfected with SLC26A9 siRNA or sterile water. The cells were assayed for 36 Chloride uptake, with and without the presence of DIDS (inhibitor of basolateral chloride/bicarbonate exchanger) 48 hours after transfection. For each transfection, RNA was assessed by PCR to verify knockdown of SLC26A9.

Results: Preliminary experiments confirmed that A549 cells express SLC26A9 and exhibited activity consistent with chloride/bicarbonate transport activity. There is a reduction in mRNA of SLC26A9 in siRNA+ transfected A549 cells compared to siRNA-transfected cells. However there was no significant reduction in the DIDS-inhibitable 36 Chloride uptake in the siRNA+ compared to the siRNA-transfected cells.

Conclusions: SLC26A9 knock-down by siRNA does not significantly affect 36 Chloride uptake in A549 cells. A recent publication (Bertrand CA et al. J Gen Physiol 2009; 133:421) demonstrated interaction between SLC26A9 and CFTR, and that functional CFTR is needed for SLC26A9 to function as a chloride transporter in human bronchial epithelial cells. As A549 cells do not express CFTR, further studies following transfection of CFTR into A549 cells may help elucidate the role of SLC26A9 in chloride transport in small airway epithelium. The efficiency of the mucociliary clearance (MCC) system depends critically on the height and properties of a thin airway surface liquid (ASL) layer that covers the respiratory tract. Compelling evidence shows that extracellular nucleotides in the ASL play an important role in the control of ASL volume and MCC function. Recently, we demonstrated that ATP release from alveolar A549 and bronchial 16HBE14oepithelial cells is a mechanosensitive, Ca 2+ -dependent process that is strongly amplified by the synergistic autocrine/paracrine actions of co-released adenosine and uridine nucleotides. Studies of the purinergic regulation of ASL volume have been hampered, however, by the lack of appropriate methods that allow the direct monitoring of ASL in real-time. We have recently developed a side-view fluorescence microscopy technique that allows the direct monitoring of the ASL on cell monolayers cultured at air-liquid interface. The aims of this study were to test our method with different airway epithelial cell models, including CF and non-CF cells, and to optimize the image analysis method for ASL volume quantification. During side-view imaging experiments, the basolateral side of the cell monolayer is perfused with physiological solution (37∞C), while the apical side is exposed to airflow (37∞C, humidity ~90%), mimicking the lung micro-environment. The Alexa 488 dextran-labelled ASL is observed at different angles, ranging from 0∞ to 90∞, permitting the direct observation of ASL distribution and thickness. Results with 16HBE14o -, A549 and Calu-3 epithelial cell models indicate that ASL isn't homogenously spread over the cell monolayer and that its local thickness varies significantly due to its accumulation in crypts between cells. Consequently, ASL volume per surface area, e.g. in µl/cm 2 , provides a better description than ASL height at any point. Similar ASL distribution was observed with both a bronchial epithelial CFBE cell line that expresses the ∆F508 mutation, and the wild-type CFTR-corrected CFBE line. To quantify ASL volume, the most convenient approach was to observe the apical surface of the cell monolayer at 45∞. In this configuration, only a narrow bright fluorescent strip remains in focus, due to the depth of field effect. The variable width of the strip is directly related to local ASL height, from which the ASL cross-sectional area, average thickness and volume per surface area was estimated using the Metavue image analysis software (Molecular Devices, PA). Experiments with Calu-3 cell cultures showed that a thapsigargin-induced (1 µM) intracellular calcium elevation transiently increased the ASL volume from a basal level of 0.7 µl/cm 2 , to 0.95 µl/cm 2 within two minutes, and then returned to its baseline level after 15 minutes. The experiments demonstrate the feasibility of our imaging technique to quantify the volume and distribution of ASL on cell culture surfaces, and to study the changes induced by physiologically-relevant stimuli in real-time.

(Supported by the Canadian Cystic Fibrosis Foundation (CCFF). D.D. is a recipient of a CCFF studentship.) The airway epithelium is lined by a small amount of airway surface liquid (ASL). Studies of exhaled breath condensate suggest that glucose con-centration in this fluid is one tenth that in plasma. During hyperglycemia, ASL glucose concentration increases while the gradient of plasma to ASL glucose concentration is maintained. Interestingly, in normoglycemic humans with cystic fibrosis, this gradient is disrupted resulting in an abnormally high glucose concentration in ASL. We studied this glucose concentration gradient in vitro, and showed that well-differentiated HAE cultures generate a transepithelial glucose gradient with low ASL glucose concentration. We investigated the mechanisms that generate the transepithelial glucose gradient. We found that HAE do not show electrogenic glucose transport, suggesting absence of sodium-glucose cotransporters (SGLT). Immunofluorescence showed that the facilitated diffusion transporter GLUT-1 is expressed in the basolateral membrane and GLUT-10 is expressed in the apical membrane. Absence of SGLT expression was confirmed by immunofluorescence. In bidirectional flux assays, we found that HAE have low paracellular permeability to L-glucose. Moreover, 2deoxyglucose fluxes favored basal to apical transport. Finally, we studied the physiological significance of the transepithelial glucose gradient. We found that increasing ASL glucose results in increased growth and infection rates with P. aeruginosa in vitro and in vivo. These data suggest that increased plasma glucose concentration or disruption of the paracellular barrier could result in increased glucose in the ASL and increased risk of pulmonary infection. In the CF lung, an increased ASL concentration of glucose and other potential carbon sources could be an additional factor that promotes airway colonization by bacterial pathogens. Hill, D.B. 1 ; Lindley, B. 2 ; Forest, M.G. 3 ; Superfine, R. 4 ; Kesimer, M. 5 ; Boucher, R.C. 1 ; Sheehan, J.K. 5 1. Cystic Fibrosis Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; 2. Mathematics, University of South Carolina, Columbia, SC, USA; 3. Applied Math, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; 4. Physics and Astronomy, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; 5. Biochemisty and Biophysics, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA The clearance of mucus from the respiratory tract is a vital component of innate immunity and is essential for respiratory health. In cystic fibrosis, it has been proposed that the breakdown of mucociliary clearance, which leads to mucus plaques and an increased rate of infection, is linked to the thickening of the mucus layer. It is the goal of this project to establish the relationship between the viscoelastic properties of mucus and its rate of clearance. Well-ciliated human bronchial epithelial cultures that demonstrate coordinated mucus clearance serve as our model system for measuring mucus clearance. Mucus layers and mucus simulants, laced with micron sized tracer particles, are loaded onto well washed cultures and their motion was measured at multiple heights through the mucus layer at high frame rates.

The results indicate a complex relationship between the viscoelastic properties of the mucus layer and its mean transport. The mucus transport rate initially increases as the viscosity of the layer increases before reaching a threshold viscosity at which mean transport decreases. Further, results indicate that while average rate of transport is constant as a function of height over the time course of the experiments, oscillations in mean transport velocity are present. These oscillations, which occur at a congruent frequency to the underlying ciliary beat, decay with the distance from the mucus ciliary interface, and can be used to define the viscoelastic properties of the fluid, as well as the strains and strain rates imparted on the mucus layer by cilia.

Our attempt to understand how the viscoelastic properties of mucus may change under different transport conditions has lead to the development of a new device, a Micro-Parallel Plate Rheometer. At the core of this device is a piezoelectric stage that oscillates one plate of the device at a set frequency and amplitude relative to a second fixed plate. The gap between plates is small, between 50 -500 µm, and when the driven oscillations are on the same order as those seen in mucus transport, the device is able to successfully mimic the oscillatory component of mucus transport. Under these controlled conditions, we are able to accurately measure the bulk viscoelastic properties of only 10 -20 µL of a complex fluid like mucus, compared to the 1 -2 mL required for standard cone and plate rheometry. Our results show shear thinning in mucus and mucus simulants over similar strain rates to those seen in our cell culture experiments.

Swaminathan, V. 1 ; Hill, D.B. 2 ; O'Brien, E. 3 ; Superfine, R. 3 1. Curriculum in Applied Sciences and Engineering, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; 2. Cystic Fibrosis/Pulmonary Research and Treatment Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; 3. Departhment of Physics and Astronomy, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA Determining the force generated by individual human lung epithelial cilia is critical for predictive modeling of many physiological functions including the ability of cilia to clear pathogens and particulates via transport of mucus. Such a measurement also allows a reasonable calculation of the force produced by each dynein motor during the active stroke, which is of interest to understanding motor function within all cellular systems. We covered 2.8 µm magnetic beads with gold, except for a small area of streptavidin on each bead, and used biotinylated wheatgerm agglutinin to bind the beads to cilia. We recorded bead motion while we applied forces on the bead via magnetic fields generated by a 3-Dimensional Force Microscope. We found that forces up to those sufficient to reduce beat amplitude by 85% did not alter the cilia beat frequency, and thus the axoneme behaved as an internally driven system throughout its beat cycle. Forces in excess of 160 pN were required to reduce the beat amplitude below 15%. We cast these results into a simple one dimensional model of axoneme dynamics, which yields an effective internal motor force of 62pN ± 14pN, or ~ 6pN per active dynein. MaxiK channels (i.e. BKCa) are voltage and calcium dependent potassium channels constituted by a tetramer of the pore-forming alpha subunit (KCNMA1). We previously demonstrated that MaxiK channels play a role in nucleotide-stimulated chloride transport in bronchial epithelial cells differentiated at the air-liquid interface. MaxiK activity is regulated by the presence of beta regulatory subunits (KCNMB1-B4). KCNMB1 and KCNMB4 positively modulate response to calcium in MaxiK channels, while KCNMB3 and KCNMB4 are inhibitory subunits. Expression of KCNMA1 and KCNMB1-B4 was studied in normal (NBE) and cystic fibrosis bronchial epithelial (CFBE) cells from human donors, cultured and differentiated at the air−liquid interface (ALI). We found that KCNMA1 and KCNMB4 were expressed both at mRNA and protein level in NBE (from 9 donors) and CFBE (from 3 donors) cells, but the mRNA expression of KCNMB4 was higher in CFBE cells (p<0.005). On the other hand, KCNMB1 mRNA expression could not be detected in NBE or CFBE cells while the inhibitory regulatory subunits KCNMB2 and KCNMB3 appear decreased in CFBE cells compared to NBE. Interestingly, TGFbeta stimulation of cells for 20 hours caused an increase of KCNMA1 and KCNMB4 mRNA in both NBE and CFBE cells. However, only CFBE showed an increase of KCNMB1. When the MaxiK opener NS1619 was used in Ussing chamber experiments, chloride secretion was stimulated in NBE but not in CFBE cells. Pretreatment with TGFbeta elicitied some response in CFBE cells in response to NS1619. The data indicate that alpha and beta subunits of MaxiK channels are expressed in both NBE and CFBE cells, but that there seem to be differences in the regulation of of MaxiK activity between the NBE and CFBE cells. Center, UNC, Chapel Hill, NC, USA CF females exhibit an accelerated decline in lung function and reduced survival rates compared to CF males. The mechanism(s) underlying such a gender disadvantage are unknown. However, we have previously shown that physiologic levels of E2 inhibit ATP/calcium dependent airway surface liquid (ASL) homeostasis by inhibiting calcium signaling (1) . This is predicted to adversely affect airway epithelial function and induce ASL volume depletion and mucus stasis, leading us to hypothesize that E2-mediated inhibition of airways calcium signaling contributes to the gender disadvantage seen in CF females. Using a CF nasal cell line (JME cells) that endogenously expresses estrogen receptor alpha and stromal interacting molecule 1 (STIM1), we measured changes in intracellular calcium levels by imaging Fura2 as described previously (1) . In the absence of extracellular calcium, 10 µM ATP caused a change in Fura2 emission ratio (F 340/380 ) of 0.63 ± 0.11 (n=12). The subsequent reintroduction of extracellular calcium caused an additional increase in F 340/380 of 0.39 ± 0.07 (n=12). A brief (15 min) pretreatment with 10 nM E2 had no effect on the ATP-dependent change in F 340/380 under calcium-free conditions (n=12). However, E2 significantly reduced the change in F 340/380 seen in the presence of extracellular calcium from 0.39 ± 0.07 to 0.1 ± 0.04, suggesting that E2 specifically inhibits calcium influx (n=6). ATP-stimulated P2Y2 receptors raise intracellular IP3 which stimulates calcium release from the endoplasmic reticulum. This then triggers STIM1 to aggregate and activate calcium influx. We next tested the hypothesis that E2 prevents STIM1 from initiating calcium influx. Knockdown of STIM1 by >90% with shRNA abolished calcium influx (n=6), confirming that STIM1 is indeed required to initiate calcium influx. We then transfected cfpSTIM1 and yfpSTIM1 into JME cells and measured STIM1 aggregation using Forster resonance energy transfer (FRET). 10 µM ATP addition increased STIM1 FRET from 14 ± 2.6% under basal conditions to 33 ± 4.6% (n=6; p<0.05). A 10 min preincubation with 10 nM E2 abolished this increase in FRET (post-ATP FRET with E2 was 12%; n=6). Further qualitative analysis revealed that ATP exposure caused the appearance of STIM1 punctae and E2 alone did not change STIM1 localization. However, E2 and ATP together caused STIM1 to be mislocalized and prevented the formation of puncta. Thus, we propose that E2 exerts its negative effects on airway epithelia by preventing STIM1 from aggregating to initiate calcium influx. The nature of the interaction between E2 and STIM1 remains to be determined. The study of mucociliary clearance in vivo would be significantly advanced by the development of a high-speed imaging technology for visualizing the respiratory mucosa at the cellular level, and could yield new insights into disease pathogenesis. Spectral domain optical coherence tomography (SDOCT) is an emerging technique capable of providing reflectance images with subcellular resolution at video rates.

Methods: We have developed an SDOCT system with an axial resolution of 1.8 µm and a transverse resolution of 2 µm. Cross-sectional images were acquired at a speed of 40 frames per second with 512 axial lines (reflectance as a function of depth) per image. We additionally obtained three-dimensional images by scanning the sample in two dimensions. The three-dimensional field of view was 200 (H) x 600 (W) x 200 (D) µm. Three-dimensional images of formalin-fixed, sectioned porcine bronchial tissue segments were obtained immediately following prosection ex vivo. Images were compared to H&E stained histology at corresponding sites. Pilot images on fixed human airways from CF and non-CF individuals were also successfully acquired.

A total of 400 cross-sectional images were acquired and digitally stored within 10 sec. Cilia were easily identified in the image and individual epithelial cells were also visible, including cellular morphology. Three-dimensional images showed gland ducts containing mucus.

Conclusions: Our results demonstrate the potential of SDOCT to provide video-rate subcelluar imaging of pulmonary airways without administration of a contrast medium. The future development of a probe for in vivo monitoring of mucociliary transport, cilia beating frequency, gland function, and ASL depth in vivo could provide new avenues for improving our understanding of respiratory mucosal pathophysiology in real-time, and enable longitudinal assessment of the response to novel CFTR modulators and other agents that restore normal ion transport.

Cross-sectional image of swine trachea. C: cilia; EC, epithelial cell; GD: gland duct. Purpose: The abundance of human leukocyte elastase (HLE) in airways contributes to respiratory disease in patients with cystic fibrosis (CF). There is a need for blood biomarkers of CF lung disease to study new therapies. Rayner et al (Respir Med 1991, 85:139-45) reported that the plasma concentration of human leukocyte elastase and alpha1 antiproteinase (pHLE-AP) complex may be a useful marker of CF lung inflammation. The purpose of this study was to determine whether pHLE-AP is a valid marker of lung disease and respiratory exacerbations in CF.

Methods: Twenty-eight individuals with CF (age 25.9 ± 1.2 y) and 47 healthy volunteers were recruited. Blood was mixed with an antiproteinase and anticoagulant cocktail, centrifuged and plasma was collected. The concentration of pHLE-AP complexes was measured using an ELISA kit (Bender MedSystems, Vienna, Austria). Correlations between values obtained from stable CF patients and forced expiratory volume in one second (FEV1) and forced vital capacity (FVC) were analyzed. pHLE-AP complexes were also compared during respiratory exacerbations and after antibiotic therapy in 10 CF patients.

Results: HLE-AP complexes were more abundant in plasma from CF patients with stable lung disease than from healthy volunteers (p < 0.005). An inverse correlation was observed between pHLE-AP complexes and FEV1 (r = 0.71, p < 0.0002) as well as FVC (r = 0.67, p < 0.001) in stable CF patients. All CF patients had increased pHLE-AP complexes during acute respiratory exacerbations and the levels returned to those observed in stable CF patients after 2 weeks of intravenous antibiotics (p < 0.001 before vs after antibiotics). Age, gender and smoking history had no effect on values.

Conclusions: These results indicate that plasma HLE-AP complexes are inversely correlated with lung function in patients with stable CF. pHLE-AP increases during respiratory exacerbations and decreases after intravenous antibiotic therapy. pHLE-AP complexes represent a simple and clinically meaningful surrogate biomarker for CF lung disease. We submit that pHLE-AP complexes may be particularly useful in clinical trials of novel therapies for CF lung disease. (MPA) . Intermittent acute viral infections have been associated with decreased lung function and disease progression. Rhinovirus (RV), which is responsible for the majority of common colds, has been detected in approximately half of CF patients with virus-related acute respiratory complaints, suggesting that RV is an important cause of CF exacerbations, as it is in patients with asthma and chronic obstructive pulmonary disease. We hypothesized that MPA infection facilitates RV binding and replication in airway epithelial cells. To test this hypothesis, we examined viral RNA levels of RV39, a major group virus, in primary normal and CF airway epithelial cells with or without MPA infection. We also examined the cytokine and interferon responses of normal and CF cells to MPA/RV39 coinfection.

Method: Primary well-differentiated normal and CF cells were infected with MPA at MOI of 0.1 or sham-infected with PBS and incubated for 24 h. After washing the cell apical surface with PBS, cells were shifted to fresh media and superinfected with intact RV39 or replication-deficient UV-irradiated RV39 and incubated for 1 h at 4∞C (to assess binding) or 33∞C for 24 h. The apical surface was washed and total RNA prepared from the cells. Basolateral medium was collected for cytokine analysis. Viral RNA (vRNA) was quantified by qPCR.

Results: vRNA levels indicated a higher viral load 1 h after infection with intact RV or UV-RV in cells pre-infected with MPA, suggesting increased binding of the virus in MPA-infected cells. After 24 h incubation, MPA/RV39-infected cells showed significantly higher vRNA levels compared to MPA/UV-RV39-, RV39-or UV-RV39-infected cells, suggesting increased replication of RV39 in MPA-infected cells. There was no difference in vRNA levels between normal and CF airway epithelial cells at either 1 or 24 h incubation. Both normal and CF cells co-infected with MPA and RV39 showed significantly increased levels of IL-8, ENA-78 and IL-6 than RV39-, UV-RV39-, MPA-or MPA/UVRV39-infected cells. Finally, MPA and RV39 co-infected cells showed significantly less mRNA expression of interferons and the interferon-induced antiviral protein, viperin.

Summary: Our results indicate that MPA infection not only increases binding of RV to airway epithelial cells, but also increases RV replication by suppressing interferon and interferon-inducible antiviral proteins.

Supported by the Cystic Fibrosis Foundation (SAJJA08G0). Airway epithelial cells (AEC) are the initial site for recognition and defense against inhaled pathogenic bacteria. Among the host defense molecules produced by these cells are the inducible antimicrobial peptides, which include β-defensins and the cathelicidin LL-37. In AEC, human β-defensin (hBD) 2 and 3 are induced upon recognition of bacteria by receptors which include the Toll-like receptors (TLRs) and CD14. We previously demonstrated that LL-37 is induced by the active form of vitamin D, 1,25(OH) 2 D 3 in cultured normal human bronchial epithelial (NHBE) cells. To examine this response further, we stimulated air-liquid interface (ALI) cultured cell lines from either normal or cystic fibrosis (CF) patients with 1,25(OH) 2 D 3 . We observed that LL-37 mRNA and protein levels increased in all cell lines when stimulated either in the basolateral medium or on the apical surface, as measured by quantitative PCR and western blot, respectively. Furthermore, antimicrobial activity was similarly increased on the apical surface of the ALI cultures in response to 1,25(OH) 2 D 3 . Analysis of the promoter region of the LL-37 gene identified a C/EBPα site adjacent to the vitamin D response element (VDRE) necessary for the induction in airway cells. To determine whether other innate immune genes were similarly induced in airway cells in response to 1,25(OH) 2 D 3 , we tested an innate immunity PCR array with cDNA from stimulated and unstimulated NHBE cells. We observed that mRNA encoding Triggering Receptor Expressed on Myeloid cells (TREM)-1, was induced over 20-fold in response to 10 -8 M 1,25(OH) 2 D 3 in ALI cultures of both CF and normal cells. The response was rapid, with a 3-6-fold induction within 3 hours, and a peak of 6-15-fold increase by 12 hours. TREM-1 is a novel pattern recognition receptor that has been demonstrated to amplify the innate immune response in monocytes. Our results demonstrate for the first time that TREM-1 gene expression is regulated in AEC, and that vitamin D can increase the innate immune defense of the airway epithelium. This suggests that vitamin D treatment of airways in patients with CF could lead to an increased innate immune defense.

Supported by a grant from the Cystic Fibrosis Foundation.

Inflammation and infection are major contributors to the ultimate morbidity and mortality in CF lung disease. Inflammation in the CF lung is excessive, but inefficient at resolving infection with pathogens such as Pseudomonas aeruginosa (P.A.). It has been shown that macrophages express Cftr (Nat Cell Biol 8:933. 2006 ) and directly contribute to the exaggerated inflammatory response in CF knockout mice (AJRCMB 40:295. 2009 ). We hypothesize that not only do myeloid cells directly contribute to the exuberant response in CF, but Cftr expression directly impacts macrophage phenotype and response to infection even prior to infection exposure. In bone marrow (BM) studies using chronic infection with P.A., not only can wild type (WT) BM decrease CF knockout mortality due to chronic infection with P.A. (from 0% survival to 54% survival) but CF BM given to WT mice significantly decreased survival of WT mice in response to P.A. chronic infection (from 55% survival to 22% survival). These observations suggest that BM-derived cells have important contributions to CF. BM includes myeloid progenitors macrophages and neutrophils, essential players in host innate immunity. To study the direct role of Cftr on macrophage function and activity, we generated a myeloid-specific Cftr knockout mouse using the mechanism of cre/loxP. In this system Cftr exon 10 is floxed by loxP sites resulting in excision of exon 10 in myeloid cells with cre-recombinase production regulated by the lysozyme promoter. Evaluation of myeloid specific Cftr bronchoalveolar lavage (BAL) cell pellets showed a significant decrease in peroxisome proliferator activator receptor gamma (PPARγ) expression relative to WT controls (-2.06±0.7 fold less PPARγ expression compared to WT, p=0.05). PPARγ expression was also decreased in BAL cell pellets of Cftr null animals (B6.129P-2-Cftr tm1Unc ,-2.88±0.11 fold less PPARγ expression compared to WT, p<0.001), and CF patient sputum cell pellets (-21.50±7.50 fold less PPARγ gene expression compared to healthy control, p<0.03). The absence of sufficient PPARγ may contribute to the overt inflammatory response found in CF (J Cyst Fibros 7:68. 2008 ). To determine if decreased myeloid specific PPARγ could contribute to the enhanced inflammation and infection in vivo, we evaluated the ability of the myeloid specific PPARγ knockout (WT Cftr) to resolve chronic infection with P.A. Our results showed that 75±7% of the myeloid-specific PPARγ deficient animals died in response to chronic infection with P.A. compared to 100% survival of WT controls. These data suggest an essential role of myeloid PPARγ in resolving chronic lung infection with P.A. To evaluate if the deficiency of PPARγ was related to loss of Cftr function, we induced PPARγ expression in the macrophage cell line RAW 264.7 in the presence or absence of Cftr inhibitor I172 (I172, 10 µg/ml). I172 inhibited PPARγ expression in LPS-stimulated cultures (-2.1±0.7 fold less PPARγ compared to no I172). These data suggest that Cftr deficient myeloid cells have altered expression of PPARγ, resulting in a phenotype which can directly contribute to the excessive inflammatory response in CF.

Funded Pulmonary infection leads to chronic inflammation and irreversible loss of lung function over time. Pseudomonas aeruginosa is arguably the most important pathogen in cystic fibrosis (CF) patients leading to this loss of lung function. Currently our therapeutic strategies to treat P. aeruginosa infection focus on antimicrobials and airway clearance; however, these treatments are only partially effective because they cannot eliminate the inciting pathogen. An alternative therapeutic approach is to target the host inflammatory response, neutralizing ineffective or overtly inflammatory components, while preserving functional immunity that prevents acute dissemination of P. aeruginosa. Currently the inflammation-targeted therapies are not widely used due to difficulties with monitoring (NSAIDS) or significant side effects (steroids). Identification and characterization of other inflammatory targets is critical to developing new therapies. Our prior work identified IL-17 as critical to driving neutrophil recruitment in an agarose bead model of P. aeruginosa infection (Dubin PJ and Kolls JK, Am J Physiol Lung Cell Mol Physiol 2007;292:L519-28). Our subsequent work suggested that IL-23, an upstream cytokine to IL-17, might also be critical to neutrophil recruitment through actions which were dependent-and independent of IL-17. We hypothesized that: IL-23 mediates neutrophil recruitment seen in P.aeruginosa airway infection through interactions with IL-1β and through the downstream activation of Th17 cells. Our data from in vitro and acute in vivo murine models of pulmonary infection with P. aeruginosa demonstrate that IL-23 is a critical innate immune mediator of neutrophil recruitment through its interactions with IL-1β to precipitate the production of inflammatory cytokines, chemokines and growth factors by macrophages; the IL-23/IL-1β effects seen are synergistic. We have also demonstrated the IL-23induced production of IL-17 by γδ T cells and a specific Th17 response during the adaptive immune response in P. aeruginosa infection, using both "acute" and "chronic" (agarose bead) models of P. aerguinosa infection. This response is characterized by the production of IL-17, IL-17F and IL-22. In these studies, neutralization of the IL-23 and Th17 responses did not lead to worsened pulmonary infection or dissemination of infection. These studies demonstrate the importance of IL-23, IL-17 and Th17 cells to P. aeruginosa-induced inflammation. Because these inflammatory mediators are critical to neutrophil recruitment, dispensable for P. aeruginosa containment and key check-points in the inflammatory response, they are promising targets for anti-inflammatory therapy in CF.

Tan, H. 1, 3 ; Regamey, N. 2 ; Hilliard, T. 1,2 ; Alton, E. 2 ; Bush, A. 1 ; Lloyd, C.M. 3 ; Davies, J. 1, 2 1. Respiratory Paediatrics, Royal Brompton Hospital, London, United Kingdom; 2. Gene Therapy, Imperial College, London, United Kingdom; 3. Leukocyte Biology, Imperial College, London, United Kingdom We have previously reported significant features of airway remodeling and submucosal inflammation in airway wall biopsies of children with CF. In contrast to the neutrophilic inflammation in the airway lumen, lymphocytes predominate in the airway wall. The phenotype and roles of these lymphocytes have not been explored. It is postulated that the Th17 pathway could play a pivotal role in CF as IL17, secreted by Th17 cells, is proinflammatory, and regulates granulopoiesis as well as neutrophil recruitment. Raised levels of IL17 have been observed in the BAL of patients with CF and non-CF bronchiectasis (Bx). The aim of this study was to characterize the lymphocytes in the submucosal area of endobronchial biopsies of patients with CF, non-CF Bx and healthy controls, staining for CD4 (T helper cells), CD8 (cytotoxic T cells) and IL17.

Immunohistochemistry was performed on endobronchial biopsies obtained from 4 groups of children: newly diagnosed CF (n=4, mean[SD]age 2.8 [1.3] years), established CF (n=24, 9[4.3] years), non-CF Bx, including PCD (n=18, 10[3]years) and controls without airway disease (n=8, 7.6 [5.4] years). BAL samples were sent for bacterial culture, total and differential cell counts, and supernatant was stored for cytokine analysis.

There was a significant difference in CD4 counts between groups (KW p=0.03), with the highest counts observed in patients with non-CF Bx (median [IQR] cells/mm2 232 ) and established (195[128-268] ) appeared increased compared to controls, but the small sample was underpowered; further samples are being analysed.

Within patients, the IL17 cell count did not correlate with levels of inflammatory markers in BALF. This may be due to the fact that a single small biopsy is not representative of the whole of the lower airway. Alternatively, inflammation in CF may be compartmentalized with true differences existing between the airway wall and lumen. There was also no difference in IL17 cell counts between patients with positive or negative bacterial cultures.

Significant differences between disease groups have been found in T helper (CD4+) cells but not cytotoxic (CD8+) T cells. Both established CF and non-CF Bx samples had higher numbers of IL17+ cells, suggesting that this is not a CFTR-specific phenomenon, but relates to chronic inflammation. IL17 cell counts were higher than CD4 counts in sequential sections of biopsies in 45% of samples, suggesting that cells other than those conventionally described as Th17 (CD4+) may produce IL17. Experiments are underway to double stain for phenotype and cytokine to determine the cellular origin of IL17.

Li, T. 1 ; Lin, T. 2 ; Cowley, E. 1 1. Physiology and Biophysics, Dalhousie University, Halifax, NS, Canada; 2. Microbiology and Immunology, Dalhousie University, Halifax, NS, Canada S100A7, also called psoriasin, is a low molecular weight, Ca 2+ -binding protein, originally described in psoriatic skin lesions, and which possesses antimicrobial activity in addition to acting as a chemotactic agent for T lymphocytes and neutrophils. Airway epithelial cells produce a number of powerful antimicrobial agents, often in common with the skin, since both skin and the airways ultimately function as barriers and come into contact with significant amounts of microflora. Therefore, we wished to investigate whether airway epithelial cells expressed psoriasin, and hypothesized that if present, its expression might be affected by physiological factors pertinent to the CF lung, for example oxidant stress and pro-inflammatory cytokines. Using RT-PCR and immunoblotting, we determined that psoriasin is expressed at the transcript and protein levels in a number of model human airway epithelial cells lines. Comparison of the normal bronchial epithelial cell line 16HBE14o-and the CF bronchial epithelial cell line CFBE41o-(∆F508/∆F508) determined that significantly less psoriasin was expressed, both at the mRNA and protein level, in the CFBE41o-cells compared to the normal 16HBE14o-. Using quantitative PCR, we found that treatment with the oxidant stressor t-butylhydroperoxide (t-BOOH) at 500 µM for 24 hours significantly reduced psoriasin gene expression in the 16HBE14o-cells, while 150 µM t-BOOH (which was the highest dose of t-BOOH we could administer without causing cell death) had no effect in CFBE41o-cells. Next, we investigated psoriasin expression in response to IL-22: 16HBE14o-cells treated with IL-22 (40 ng/ml) showed a significant increase in psoriasin gene expression, while no change was apparent in the CFBE41o-cells. Finally, since high levels of pro-inflammatory cytokines have been measured in the CF airway, we wished to investigate whether exposure to IL-1β, TNF-α or IFN-γ would have any effect. Exposure to IL-1β for 24 hours had no effect on psoriasin gene expression in the 16HBE14o-cells, but significantly increased expression in CFBE41o-cells. Treatment with either TNF-α or IFN-γ (again for 24 hours) significantly increased psoriasin gene expression in both cell lines, though the increases in the 16HBE14o-cells were less pronounced than in the CFBE41o-cells. Presently, studies using the mucoid Pseudomonas aeruginosa strain 8821 (a gift from A. Charkrabarty, University of Illinois) are underway to determine whether the enhanced expression of psoriasin seen in response to IL-22 treatment is associated with increased bacterial internalization and killing.

In conclusion, psoriasin is expressed in normal and CF airway epithelial cells, though at a much reduced level in the CF cells. We believe psoriasin may play a role in maintaining a sterile environment in normal airways, while exposure to pro-inflammatory mediators and oxidant stress could result in an enhanced expression, possibly representing an adaptive host defence response to combat deleterious bacterial products. Supported by the Canadian Cystic Fibrosis Foundation.

Kurlandsky, L.E.; Anbar, R.D.; Soultan, Z.N. Pediatrics, SUNY Upstate Medical University, Syracuse, NY, USA Background: Endogenous production of carbon monoxide (CO) as a byproduct of the breakdown of heme by heme oxygenase (HO-1) has been demonstrated to have antioxidant and antiinflammatory functions. Its production is increased in patients with cystic fibrosis (CF) and found in increased concentrations in the exhaled breath of these patients. Because exhaled CO correlates with serum carboxyhemoglobin (HbCO), we sought to evaluate the clinical usefulness of measuring HbCO in patients with CF with a recently available, non-invasive instrument that measures serum HbCO transcutaneously.

Methods: A transcutaneous CO-Oximeter (Rad-57, Masimo Corp., Irvine, CA) measures differences in light absorption between HbCO (SpCO), methemoglobin, and oxygenated and unoxygenated hemoglobin. This instrument is available at our Pediatric Pulmonary and Cystic Fibrosis Center to evaluate HbCO resulting from cigarette smoke or other environmental CO exposure.

Results: Measurements from a convenience sample of 22 patients with CF who do not smoke (mean age 11 years (range, [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] , and 36% male), showed an average SpCO of 5.9 +/-3.4% (range, 0-12). Their average FEV1 was 76 +/-27% of predicted for height and age (range, 18-113%). The correlation coefficient between level of SpCO and FEV1 was 0.87 (see Figure) . Twenty three controls composed of pediatric residents and clinic staff who do not smoke, had an average SpCO of 0.08% +/-0.28% (range, 0-1).

Conclusions: This preliminary series of observations supports reports of increased CO in the exhaled breath of patients with CF. The positive correlation of HbCO, measured by SpCO, and FEV1 suggests that HbCO is not only a marker of inflammatory and oxidative stress but may be a measure of a beneficial adaptive response. This concept is supported by an in vitro study of a CF airway epithelial cell line that showed overexpression of HO-1, which results in the release of CO, led to protection against Pseudomonas aeruginosa-induced injury and apoptosis (Zhou et al. Am J Respir Crit Care Med 2004; 170:633-640 CF lung disease is characterized by an exaggerated inflammatory response and susceptibility to microbial colonization, particularly with Pseudomonas aeruginosa (Pa). Overall in CF, host immune response does not seem to be adequate to eradicate Pa from the respiratory tract. CFTR is expressed in a variety of cells of the immune system, including dendritic cells (DC). Given the critical role of lung DC to initiate adaptive immune response against pathogen infections of the respiratory tract, the present study characterizes lung DC subsets including conventional DC (cDC) and plasmacytoid DC (pDC) from CF mice and their response to pulmonary infection with Pa. DC were isolated from lung suspensions of naive CF mice (Cftr tm1UNC knockout mice) and WT mice (C57/BL6 mice) or 6 days following intranasal administration of Pa (1x10 5 cfu/mouse). The lung DC subsets were analyzed as cDC (CD11b high CD11c high cells) or pDC (CD11b -CD11c + B220 + cells), and their activation and maturation profiles by surface expression of CD40, CD80, CD86, MHCII, and PCDA1 by flow cytometry. Both pulmonary DC subsets in naive CF mice were lower compared to WT controls (cDC: 1.6 ± 0.1% in CF mice, 2.9 ± 0.1% in WT mice, p<0.01; and pDC: 0.5 ± 0.1% in CF mice, 0.9 ± 0.1% pDC in WT mice, p< 0.01). The DC subsets from the CF mice showed less maturation and activation compared to the WT controls at baseline, with lower expression of CD40, CD80, CD86, and MHCII (p<0.05, all comparisons). Mixed lymphocyte reaction (MLR) was performed to assess the allostimulatory capacity of DC. The proliferation index of allogeneic T cells stimulated by CF DC was 1.7-fold lower compared to the WT controls (p<0.01). Following infection with Pa, the percentage of cDC decreased 1.4 ± 0.1-fold in WT mice compared to 0.9 ± 0.1-fold in CF mice (p<0.05); and the pDC population decreased by 1.2 ± 0.1-fold in the WT mice and did not change in the CF mice (p<0.05). After Pa infection, cDC from CF mice showed up-regulation of CD40 (p<0.05), and MHCII expression was increased in cDC from both WT and CF mice. The proliferation index of allogeneic T cells stimulated by CF DC or WT DC following infection with Pa was comparably increased 2-fold. These data demonstrate that lung DC from CF mice show baseline differences of activation and maturation profiles and allostimulatory capacity, and an impaired response upon infection with Pa. This may be an important mechanism underlying the increased susceptibility of Pa infection in CF lung disease.

Supported by NHLBI/NIH R21 HL77557-02 and Cystic Fibrosis Foundation XU09F0.

Anil, N. 1 ; Singh, M. 2 ; Rajwanshi, A. 3 ; Vohra, H. 4 1. PGIMER, Chandigarh, India; 2. PGIMER, Chandigarh, India; 3. PGIMER, Chandigarh, India; 4. PGIMER, Chandigarh, India Background: Persistent airway inflammation is a hallmark in disease pathogenesis of cystic fibrosis (CF) lung disease. Nitric oxide is an important mediator in this inflammatory process. Low levels of exhaled nitric oxide (eNO) have previously been reported in CF.

Methods:

The aim of the current study was to compare the levels of nitrites and PMN% in induced sputum of children with CF (n =15) with those in healthy controls (n =10). Furthermore association between nitrite levels and lung function was also estimated. Nitrites were estimated based on Greiss assay; lung function was ascertained by spirometry performed according to standard guidelines.

Results: We observed a high level of sputum nitrites in children with CF (184+ 10.07) µM/l versus controls (56.4+ 5.7 µM/l) p<0.001. A high percentage of PMN in children with CF (78%) versus controls (28%) p<0.01 was observed. A significant negative correlation between induced sputum nitrites and FEV1 (r = -0.716, p<0.01), and between PMN% and FEV1 (r= -0.528, p<0.05) in children with CF was observed.

Conclusion: Induced sputum nitrites reflect the airway inflammation in the lower respiratory tract in CF, furthermore the negative correlation between sputum nitrites and lung function could serve as a crucial non-invasive indicator of the diseased state. The "gender gap" in cystic fibrosis (CF) is well documented -females have more aggressive disease, poorer lung function and earlier Pseudomonas spp colonisation (1) (2) (3) . Toll-like receptors (TLRs) are transmembrane pattern recognition receptors that respond to microbial and pro-inflammatory stimuli. TLRs are expressed by airway epithelium (4) and the CF lung is a milieu potentially rich in TLR agonists. Little is known about the inflammatory effect of the female sex hormone estrogen within the CF lung. We evaluated the interaction of 17β-estradiol (E2) with TLRs on CF airway epithelium.

Methods: Studies were performed using CF (CFTE29o-and CFBE41o-) and non-CF cell airway epithelial cell lineages (16HBE14o-) grown in monolayers or as polarized cultures. qRT-PCR was used to quantify estrogen receptor (ER) expression (alpha and beta). Laser scanning cytometry quantified changes in TLR expression on exposure to E2. ELISA of cell supernatants assessed the effects of E2 on IL-8 expression in response to TLR and other pro-inflammatory agonists (PAM3 (TLR2/1), POLYIC (TLR3), LPS (TLR4), uCpG DNA (TLR9), CF bronchoalveolar lavage fluid (BALF) and 5% Pseudomonas-conditioned medium (PCM)). The expression profile of 174 cytokines +/-E2 following stimulation with CF BALF was investigated by cytokine antibody array.

Results: ER-α and β are both expressed in CF and non-CF airway epithelia although ER-β expression is proportionally higher (2-to 6-fold) in CF. Exposure to 10nM E2 upregulates expression of TLR 1, . In response to TLR agonists, CF BALF and 5% PCM IL-8 expression was significantly increased in CFTE29o-, CFBE41o-and 16HBE14o-cells (all p<0.05). In polarised CFBE41o-cells IL-8 was released apically rather than basolaterally (p<0.001). Following exposure to E2 (1-10nM), agonist-induced IL-8 production was attenuated in a dosedependent fashion. E2 (10nM) consistently impaired TLR agonist, PCM and CF BALF-induced IL-8 in all cell types by 41-73%. E2 had stimulatory and inhibitory effects on the basal and CF BALF-induced expression of multiple cytokines.

Conclusion: ER-β is the major ER expressed by airway epithelial cells and is proportionally higher in CF versus non-CF cells. At physiological concentrations E2 can upregulate TLR1, -3 and -4 surface expression on CF airway epithelial cells but shows a dose-dependent inhibition of IL-8 secre-tion in both CF and non-CF airway epithelial cells in response to TLR1-6 & 9 agonists. E2 regulates the expression of multiple cytokines in CF airway epithelial cells.

References : Clinic 1, Rigshospitalet, Copenhagen, Denmark Background: Concentration of nitric oxide (NO) in healthy sinus exceeds that needed for antibacterial effects in vitro. In CF, nasal NO (nNO) is lower than in healthy controls. Chronic sinusitis is a common problem in CF patients. Functional Endoscopic Sinus Surgery (FESS) is a possible symptomatic treatment, and can play a role in eradication of occult bacterial focus. Level of nNO might be influenced by FESS, but neither influence of intravenous (IV) anti-Pseudomonas treatment nor natural fluctuations in nNO in CF patients without chronic sinusitis is known. Data presented at NACFC 2008 (de Winter-de Groot. Ped Pulmonol 2008;S31:377) showed significantly increased nNO after FESS.

Aim: To assess whether nNO levels increase in CF patients with chronic sinusitis after FESS or differs during IV anti-Pseudomonas treatment and compared to CF controls without chronic sinusitis.

Methods: nNO was measured in 15 CF patients (median age 14, range 8-23 yrs), at baseline and subsequently monthly for 4 months after FESS. To compare results we measured nNO in 7 patients (median age 14, range 9-25 yrs) receiving a 14 day course of IV antibiotics for Pseudomonas aeruginosa, and 7 random control CF patients (median age 17, range 7-24 yrs), without chronic sinusitis. Changes in nNO were compared to baseline for each group (paired Wilcoxon-test) and the different groups were compared using ANOVA analysis.

Results: Baseline nNO (geometric mean and 95%CI) did not differ significantly between groups and there were no significant change in any of the three groups at 1, 2, 3 or 4 months compared to baseline. Comparing the groups, there were no significant differences in nNO levels, except at the 1 month point, where the FESS group had a significantly lower level of nNO (206ppb) than both IV (512 ppb) and control (662 ppb) groups, and after 3 months, where the IV group (724ppb) had a significantly higher level than the FESS (330 ppb) and control (320 ppb) groups ( Figure 1 ).

Conclusion: There were no significant short or long term changes in nNO following FESS operation in CF patients and nNO seems to be unaffected by intravenous antibiotic course and without significant natural fluctuations during time. Background: Proinflammatory cytokine dysregulation was proposed as a cause of pulmonary inflammation in cystic fibrosis (CF). Disproportionate concentrations of proinflammatory cytokines have been recorded in CF bronchial samples, including elevated levels of the potent neutrophil chemoattractant interleukin (IL)-8 /CXCL8 and contrasting dramatically diminished levels of the IFN-γ inducing factor, IL-18. It has previously been shown that glycosaminoglycan (GAG) matrices increase the half-life of IL-8 at sites of inflammation, but why reduced levels of IL-18 are associated with CF lung disease is unclear.

Aim: The aim of this project was to compare IL-8 and IL-18, for their relative stability, activity and interaction with glycosaminoglycans (GAGs) in the lungs of CF patients.

Results: In total contrast to measured levels of IL-8, diminished concentrations of IL-18 were detected in CF bronchoalveolar lavage fluid. In apparent contradiction the pro-and mature forms of IL-18 were expressed and produced equally by CF and control epithelial cells and monocytes.

Biacore studies were designed to investigate the ability of IL-18 to bind GAG matrices. Comparative statistical analysis was performed to monitor the effect of pH and results revealed a 102% increase in binding of IL-18 to GAGs at pH 6.3 compared to pH 7.3 (P = 0.01). Surfaces coated with GAGs at a concentration of 30 µg/ml were capable of binding approximately 62 ± 6.8 pg IL-18/cm 2 . However, exposure of GAG-coated surfaces to IL-8, added either simultaneously or one hour after IL-18, competitively reduced and displaced the level of detectable IL-18 by 57% (P = 0.002) and 32% (P = 0.02), respectively. In addition, competitive displacement of IL-18 from these anionic matrices by IL-8 rendered the cytokine susceptible to rapid proteolytic degradation by neutrophil elastase.

Interestingly, high concentrations of NaCl in vitro engendered release of IL-8 from GAGs exposing the cytokine susceptible to elastase-mediated proteolysis. To determine if this could happen in vivo, sputum was collected from CF patients before and after nebulisation with 7% (w/v) hypertonic saline. Sputum collected after nebulisation illustrated lower levels of IL-8 and decreased neutrophil chemotactic activity.

Conclusion: A novel mechanism has been identified highlighting the potential of IL-8 to determine the fate of other cytokines, including IL-18 within the CF lung, consistent with the inflammatory status of the CF lung disease. Hypertonic saline treatment lowers IL-8 levels in the lung and therefore may aid inflammatory resolution. Interleukin-8 (IL-8) is the principal chemoattracting agent of neutrophildominated inflammation, which causes progressive lung tissue damage in cystic fibrosis (CF). CF respiratory epithelium secretes elevated amounts of IL-8 either spontaneously or in response to pro-inflammatory cytokines, such as tumor necrosis factor (TNF)-α. Ibuprofen, a non-steroid anti-inflammatory drug, and corticosteroids are clinically beneficial in CF, but concerns over adverse effects have limited their use. Corticosteroids, such as dexam-ethasone, are known negative regulators of IL-8 in respiratory epithelial cells. It is not clear whether ibuprofen exerts similar effects on IL-8 production in respiratory epithelial cells. We examined the effects of clinically relevant concentrations of ibuprofen (480 µM) and dexamethasone (5 nM) on nuclear factor κB (NFκB, a transcription factor up-regulating IL-8) activation and IL-8 production in a TNF-α (10 ng/ml) stimulated CFTE29o-cell line, in order to compare anti-inflammatory mechanisms. CFTE29o-cells, expressing ∆F508 mutated CFTR, were pre-incubated for 30 minutes (±) ibuprofen or dexamethasone, and stimulated for 1 hour with TNF-α. NFκB transcriptional activity, as well as IL-8 mRNA levels and protein expression were assayed.

Results: TNF-α increased NFκB transcriptional activity (p < 0.00001) and IL-8 transcription (p = 0.0002) and protein expression (p = 0.018). Ibuprofen decreased TNF-α stimulated NFκB transcriptional activity (p < 0.0001), but had no significant effect on IL-8 transcription (p = 0.99) or protein expression (p = 0.46). Dexamethasone decreased TNF-α stimulated NFκB transcriptional activity (p < 0.0001) as well as IL-8 transcription (p = 0.03) and protein expression (p = 0.006).

Conclusion: Unlike corticosteroids, ibuprofen's anti-inflammatory properties do not directly target IL-8 expression in TNF-α stimulated CF respiratory epithelial cells. Effects of dexamethasone on IL-8 production are likely NFκB-independent.

Supported by the BREATHE Initiative of the Canadian CF Foundation. CFTE29o-cells are a generous gift of Dr. D. Gruenert. Background: Cystic fibrosis (CF) is characterized by increased neutrophil elastase (NE) and reactive oxygen species in the lung. CF airways also have increased non-heme iron content. We hypothesized that NE generates reactive oxygen species in part by increasing intracellular catalytically active iron levels.

Objective: To test whether intratracheal instillation of NE into rat airways or NE exposure of human airway epithelial cells in vitro increased intracellular non-heme iron.

Methods: Male Sprague-Dawley rats (total n= 96) were intratracheally instilled with either buffer or 50 µg neutrophil elastase (0.5 mL). At 28 days after exposure, rat lungs were inflation-fixed with 10% formalin and prepared for immunohistochemistry. Iron was detected using Perls' Prussian blue stain. Ferritin was detected by immunohistochemistry using rabbit antiferritin antibody (Dako). For in vitro studies, normal human bronchial epithelial cells (NHBE) were loaded with calcein AM, a fluorescent marker that is quenched by increased non-heme intracellular iron. NHBE were cultured submerged in defined serum-free medium and exposed to NE (100 nM, 4 h) or control vehicle and calcein AM fluorescence was detected in a fluorescent plate reader. To test whether NE increased iron by increasing the pool of free extracellular iron available for uptake, we exposed cells to vehicle or Fe57-loaded ferritin (100 ng/ml) in the presence or absence of NE (100 nM, 4 h) and then following washing, lysed cells to quantitate uptake of Fe57.

Results: Rats exposed to NE had increased iron uptake in macrophages and increased ferritin in macrophages and in airway and alveolar epithelial cells. NE increased intracellular non-heme iron by calcein AM fluorescence assay. NE increased intracellular Fe57 likely by release from ferritin, increasing the pool of non-protein bound iron available for uptake.

Conclusions: NE increases intracellular iron at least in part by releasing iron bound to extracellular proteins, such as ferritin, resulting in an increase in the extracellular iron pool available for cellular uptake. Increased intracellular iron may contribute to oxidative stress in lung diseases with unopposed excessive NE.

Supported by NIH HL082504 (JAV) and HL081763 (BMF). Growing evidence suggests sphingolipids (SLs) and glycosphingolipids (GSLs) as novel targets for the treatment of pulmonary disorders such as asthma, COPD, acute lung injury and CF [Uhlig, 2008] , being involved in many biological events such as immunity, pathogen invasion and apoptosis [Hannun 2008] . Recently, the accumulation of the SL ceramide has been identified as one of the key regulators of inflammation and infection in CF airways [Teichgraber 2008] . Miglustat, an inhibitor of the synthesis of glycosphingolipids, already used for treating type I Gaucher disease, produces an anti-inflammatory effect in bronchial epithelial cells [Dechecchi 2008] . In this study we tested the effect of miglustat on murine models of lung inflammation in vivo by administrating aqueous solution of miglustat (2 mg/dose) in C57BL/6J mice by intraesophageal gavage with three administrations 72, 48 and 24 hours before intranasal instillation of lipopolysaccharide (LPS). Bronchoalveolar lavage fluid (BALF) was collected 4 h post LPS challenge. Miglustat significantly reduced the amount of neutrophils recruited in BALF by 60% whereas lymphocytes or alveolar macrophages were unchanged. These results indicate that miglustat strongly reduces neutrophil chemotaxis in vivo. To support the hypothesis that miglustat affects the imµmune response through changes in cellular ceramide levels, CF bronchial cells were treated with amitriptyline, an inhibitor of two important enzymes of the SL metabolism: acid sphingomyelinase and acid ceramidase. Amitriptyline (10 µM), added to IB3-1 and CuFi-1 cells 4 hours before infection with P. aeruginosa, significantly inhibited the expression of IL-8 mRNA by about 50%. Cellular ceramide levels were then measured in cells treated with both miglustat and amitriptyline, by FACS analysis with anticeramide antibody MAS 0020 (Glycobiotech). Ceramide staining in IB3-1 and CuFi-1 cells increased upon infection with P. aeruginosa. Both miglustat and amitriptyline significantly reduced the increase of ceramide expression induced by P. aeruginosa. Collectively these data indicate a link between the anti-inflammatory effect of miglustat and the decrease of ceramide levels and strengthen the hypothesis that SL metabolism represents a target to restore a normal inflammatory response in CF epithelium.

Supported by: Italian Cystic Fibrosis Research Foundation (#12/2008) with the contribution of "Delegazione FFC di Vicenza con le scuole vicentine." Tamanini Introduction: CF is characterized by high concentrations of IL-8 and a sustained accumulation of neutrophils in the airways, which are frequently colonized by bacterial pathogens, especially Pseudomonas aeruginosa (Pa). The search for potential therapeutic agents in CF focuses on pharmacologically mediated regulation of pro-inflammatory gene expression to reduce the adverse effects of chronic inflammation of CF lungs.

In this context, we identified 5-methoxypsoralen as a relevant molecule able to reduce the Pa-dependent transcription of IL-8 mRNA (Nicolis et al. 2008) . Here, we extended the analysis to other derivates: psoralen, 8methoxypsoralen, 4,5,8-trimethylpsoralen, angelicin and FEVR-PD12. Because psoralens have been already characterized as openers of CFTR chloride channels (Devor et al., 1997) , we wished to determine whether the compounds with the most potent anti-inflammatory effect are also able to increase CFTR-mediated ion transport.

Methods: the bronchial, submucosal non-CF cells, Calu-3, were seeded onto collagen coated, Transwell polyester membranes. After 15 days the integrity of monolayers was evaluated by measuring TER and used for experiments when the resistance exceeded 2000 Ohm/cm 2 . Monolayers were incubated for 24 hours with compounds before infection with Pa (2x10 7 CFU/ml) at 37∞C for 4 hours, total RNA isolated, reverse-transcribed and the resulting cDNA quantified by relative quantitative real-time PCR to study a panel of pro-inflammatory molecules besides IL-8 and namely: ICAM-1; GROα/γ; MIP-1α; IP10; TNF-α; IL-6; IL1β. Electrophoretic mobility shift assay (EMSA) evaluated the DNA binding activity of NF-kB, AP-1, sp-1 or GATA by Pa in cells treated with psoralens. CFTR-dependent chloride efflux in polarized Calu-3 cell monolayers was measured using the Clsensitive dye, MQAE.

Results: quantitative real-time PCR identified a selective reduction of IL-8 mRNA upon exposure of the cells to different psoralens with much higher potency using angular psoralen, FEVR-PD12. This compound demonstrated no significant alteration of DNA binding activity of the transcription factors studied in either uninfected or Pa-infected cells. Importantly, 100nM FEVR-PD12, after 24 hour of preincubation, potentiated CFTR dependent chloride efflux induced by FSK by about 40%.

Conclusions: Altogether, these results demonstrate that the psoralens are promising molecules both as modulators of the immune response and potentiators of CFTR function.

Supported by Italian Cystic Fibrosis Research Foundation (grant FFC#13/2007) with the contribution of "Associazione Veneta per la lotta alla fibrosi cistica" and Fondazione Cariverona (Bando 2005). not found to alter the rate of phagocytosis of P. aeruginosa by cultured epithelia as tested using a standard invasion assay, exposure to bacteria altered CFTR activity. Cultured 2WT2 epithelia pre-incubated for 10 min. with living P. aeruginosa (MOI 100) had 1.15X increase in forskolin-activated CFTR dependent peak iodide efflux rate. Lipopolysaccharide (LPS 50-200ug/mL) purified from P. aeruginosa extracellular surface when preincubated for 10 min. increased CFTR channel activity more potently than live bacteria (1.5X increase in forskolin-activated CFTR dependent peak iodide efflux rate), and did so in a dose dependent manner. However, shorter exposure to 100ug/mL LPS (no pre-inubation but simultaneous administration with forskolin) resulted in 0.8X reduced CFTR-dependent area under the curve of iodide efflux rate. This is the first report of a bacterial product causing a biphasic time-dependent and a dose-dependent alteration of CFTR channel activity. These data are consistent with a model in which CFTR may be removed from the plasma membrane during phagocytosis of P. aeruginosa followed by recruitment of the channel to the membrane to replace channels removed during phagocytosis. Chronic bronchial infections by Pseudomonas aeruginosa are a major problem in patients suffering from cystic fibrosis (CF), bronchiectasis or COPD. Earlier studies suggest that azithromycin (AZM) improves respiratory function in CF patients colonized with P. aeruginosa independent of the bacterial count within the lung. The underlying mechanism is debated. Previously we showed that AZM increased transepithelial electrical resistance (TER) and affected the expression of tight junction proteins in bronchial epithelia in vitro. We asked if this effect was significant to airway epithelial defense during P. aeruginosa infection.

We used an air-liquid interface model of human airway epithelia and measured TER and changes in tight junction protein expression following exposure to live P. aeruginosa (PAO1), and PAO1-∆rhl, a mutant lacking rhlA and rhlB, which encode key enzymes for rhamnolipid production.

Rhamnolipids are an established virulence factor of PAO1. In addition, epithelia were challenged with bacterial culture medium conditioned by these strains, purified rhamnolipids alone or synthetic 3O-C12-homoserine lactone (HSL), another known virulence factor of PAO1.

We found that PAO1 medium decreased TER significantly and rearranged TJ protein expression. In contrast, PAO1-∆rhl had significantly less effect on TER. Purified rhamnolipids decreased TER in a dose-dependent manner, suggesting that rhamnolipids produced by PAO1 are responsible for lowering TER in airway epithelia in vitro. Interestingly, pre-treatment of the epithelium with AZM attenuated the TER-lowering effect of rhamnolipids significantly, while decreased TER by HSL was mostly unaffected by AZM.

The data indicate that AZM maintains airway epithelial integrity during P. aeruginosa infection by counteracting rhamnolipids. The results could help explain the beneficial effects of macrolides observed in CF clinical trials. Center, UNC, Chapel Hill, NC, USA; 2. Medicine, UNC, Chapel Hill, NC, USA We have shown that human bronchial epithelial (HBE) inflammation triggers endoplasmic reticulum (ER) stress known as the unfolded protein response (UPR). UPR activation during HBE inflammation couples to stimulation of the inositol requiring enzyme 1 (IRE1), an ER enzyme that splices the mRNA of the transcription factor X-box binding protein 1 (XBP-1), a pathway required for protein secretion in many cells. Our recent data demonstrated that activation of IRE1/XBP-1 is functionally relevant for airway inflammation in vitro and in vivo, because XBP-1 mRNA splicing induces expansion of ER calcium stores, which mediate an increased calcium-dependent cytokine secretory response in inflamed airway epithelia (Martino et al. J Biol Chem 2009; 284:14904) . The present study addressed whether IRE1 is also functionally relevant for mucin production during airway inflammation. We first investigated the relative expression of the two IRE1 isoforms, α and β. IRE1α is ubiquitously expressed, whereas IRE1β is only found in gut and upper and lower airway respiratory tissues. In freshly isolated HBE, IRE1β expression was more than 30 fold higher than IRE1α expression but was absent in primary human alveolar type II cells. Studies in primary cultures of HBE revealed a positive correlation between the mRNA levels of IRE1β and MUC5B during the time course of epithelial differentiation. In mouse studies, IRE1β expression was not detected in lung parenchyma but was 8.1 and 2.3 fold higher than IRE1α expression in freshly isolated nasopharynx and trachea, tissues that exhibit higher levels of goblet cells. These findings led to the hypothesis that IRE1β might be functionally relevant to airway mucin production. This hypothesis was tested by comparing the airway responses of IRE1β -/vs. wild-type (wt) mice to ovalbumin (OVA) challenge. In wt mice, IRE1β co-localized with the Clara cell secretory protein (CC10), which expression increases following an OVA challenge as it reflects goblet cell metaplasia. Notably, IRE1β was not expressed in ciliated cells. In IRE1β -/mice, the absence of IRE1β in airway epithelia correlated with a lack of CC10 induction by OVA. Moreover, OVAincreased airway epithelial Muc5b and Muc16 production was blunted in IRE1β -/vs. wt mice. Additional evidence for a role of IRE1β related to mucin production was obtained by a decreased AB/PAS staining of goblet cells in nasopharynx epithelia from IRE1β -/mice. Finally, XBP-1 splicing resulting from ER stress induced by thapsigargin was the same in primary cultures of control vs. IRE1β -/mice, suggesting that IRE1β function is not mediated by the XBP-1 UPR pathway. These novel findings suggest that IRE1β exhibits a specific functional role in airway goblet cells, its action is distinct from IRE1α and independent from XBP-1. We conclude that IRE1β activation by airway inflammation-induced ER stress is a key functional epithelial response required for mucin production. Addressing the mechanism of IRE1β-mediated mucin production may lead to therapies targeting this pathway and benefit CF patients suffering from chronic airway inflammation and mucus overproduction. Supported by The CFF. Objective: The aim of this study was to explore the effects of macrolides azithromycin (AZM), clarithromycin (CAM) and erythromycin (EM) on molecular mechanisms involved during inflammation in cystic fibrosis (CF) bronchial epithelial cells.

Methods: Human CF bronchial epithelial (IB3-1) and CFTR-corrected bronchial epithelial cell lines (S9) were pre-treated 30 minutes with the macrolide (10 µg/ml) and TNF-α (10 ng/ml) was added for an additional 16h of culture. To confirm results, we have treated with the same protocol human bronchial gland cells (HBG) isolated from non-CF patients with or without the CFTR inhibitor Inh-172 (10 µM, 72h). IL-8 concentrations were evaluated in culture supernatants by ELISA and NF-κB and AP-1 transcriptional activity were investigated using a p65-luciferase plasmid at 4h and 8h of treatment.

Results: CAM and EM were not able to modulate TNF-α induced IL-8 secretion. Interestingly AZM significantly reduced TNF-α induced IL-8 secretion in non-CF cells (S9 or HBG cells) but not in CFTR deficient cells (IB3-1 cells and HBG cells treated with the CFTR inhibitor Inh-172). Next, we have investigated molecular mechanisms after 4h and 8h of treatment. We demonstrated that AZM significantly decreased TNF-α induced NF-κB transcriptional activity in S9 and in HBG cells but not in IB3-1 cells or in HBG cells treated by Inh-172. In S9 cells, the level of cytosolic IκBα (analyzed by Western-blotting) decreased following a treatment with TNF-α but increased after AZM + TNF-α treatment. Finally, we demonstrated that AP-1, the nuclear transcription factor induced by MAP kinases, is not significantly modulated by AZM in all cells analyzed.

Conclusion: Our study demonstrates that AZM is not able to decrease inflammation in CF bronchial epithelial cells and suggests that CFTR chloride function is necessary to anti-inflammatory effects of azithromycin in bronchial epithelial cells.

Supported in part by the French Association Vaincre La Mucoviscidose. We also thank Pfizer Laboratories for their generous gift of azithromycin. In CF lung disease neutrophilic inflammation must be targeted aggressively and effectively. Complementary to molecules correcting the basic CFTR defect, anti-inflammatory drugs have been proposed as an important strategy to ameliorate CF lung pathology. Glucocorticoids (GC) are powerful anti-inflammatory molecules but of limited benefit in CF patients [Nichols DP et al. Clin Rev Allergy Immunol 2008; 35:135] making mandatory the investigation of new candidate molecules, tailored to CF lung inflammation. The aim of this study was to compare the effect of the GC dexamethasone with miglustat [Dechecchi MC et al. J Cyst Fibros 2008; 7:555] on the host-pathogen interactions in human bronchial epithelial CF (CuFi-1) and non-CF (NuLi-1) cells exposed to P. aeruginosa. Cells were incubated overnight in the presence of dexamethasone or miglustat and then infected by P. aeruginosa for 4 hours. The mRNA expression of selected panels of genes involved in leukocyte chemotaxis and antimicrobial activity was measured by real time PCR. Both in CuFi-1 and NuLi-1 cells, P. aeruginosa significantly induced the expression of the chemokines IL-8, Groα/γ, IP10 and Rantes, the cytokines TNFα, IL-6, IL-1β, the adhesion molecule ICAM-1 and the antimicrobial betadefensins hβd2 and hβd4.

Moreover, the antimicrobial gene lactoperoxidase LPO was induced by P. aeruginosa in NuLi-1 cells. A significant inhibition of the P. aeruginosa stimulated mRNA expression of IL-8, Groα/γ, IL-6 and ICAM-1 was obtained in CuFi-1 cells by miglustat without any effect on the expression of hβd2 and hβd4. On the contrary, dexamethasone failed to reduce the inflammatory response in CuFi-1 cells, consistent with scarce efficacy of GC treatment observed in CF patients. In agreement with the results obtained in CuFi-1 cells, miglustat significantly reduced the expression of IL-8, Groα/γ, IL-6 and ICAM-1 and did not affect the expression of the antimicrobial genes, in NuLi-1 cells. Different from the results obtained in CuFi-1 cells, dexamethasone was effective in reducing the neutrophil chemotaxis in NuLi-1 cells. Moreover it also reduced the expression of the chemokine Rantes, the pro-inflammatory cytokine TNFα and the antimicrobial gene LPO. These results confirm that miglustat has an anti-inflammatory effect in bronchial epithelial cells exposed to P. aeruginosa and demonstrate no impairment of the antimicrobial activity. In respect to dexamethasone, CF bronchial cells seem to be resistant to GC treatment, which, however has some adverse effects in reducing the antimicrobial activity. Identifying the mechanisms responsible for the GC resistance in CF could provide useful information to address potential anti-inflammatory therapies. Identifying new drugs that display sufficient anti-inflammatory efficacy but no impairment of the host defence remains a key challenge.

Supported by: Italian Cystic Fibrosis Research Foundation (#12/2008) with the contribution of "Delegazione FFC di Vicenza con le scuole vicentine".

The activation of expression of pro-inflammatory genes by P. aeruginosa with bronchial epithelial cells is a central mechanism to be targeted with novel therapies. Medicinal plants are attracting a growing interest because of their potential safety, already tested in large scale applications in human diseases. The plant Nigella arvensis belongs to the botanical family of Ranunculaceae and is commonly known in native populations of North Africa for the black seeds in use as a natural remedy for over 2000 years. Ethanol extracts from N. arvensis seeds were tested in IB3-1 CF bronchial cells exposed to the P. aeruginosa laboratory strain PAO1 for 4 h. IB3-1 cells were pre-treated with 10 µg/ml of N. arvensis 20, 4 and 2 h before infection, at time of infection and 2 h after infection. Total RNA was extracted and the neutrophil chemokine inflammatory marker IL-8 transcript was quantified. N. arvensis significantly inhibited the PAO1-dependent transcription of IL-8 in IB3-1 cells at all the times tested. We focused the study on the longer treatment time (20 h) and we found that N. arvensis did not inhibit the transcription of other neutrophil chemokines such as GRO-alpha and gamma, ICAM-1 and IP-10 and of the pro-inflammatory cytokine TNFalpha, IL-6, IL-1 beta and IFN-gamma. No effect was observed on the expression of the PAO1-dependent over-expression of three antimicrobial genes, beta-defensin 2 and 4 and lactoperoxidase. In order to exclude adverse effects on cell cycle, cell growth has been studied in the IB3-1 cells pre-incubated with the extract for 24 and 48 h. No changes in cell proliferation have been observed, also at N. arvensis high concentration (200 µg/ml). To exclude the probability that the inhibition of the P. aeruginosadependent induction of IL-8 was due to an indirect anti-bacterial effect on P. aeruginosa, we performed an anti-bacterial array following the procedure for the Minimum Inhibitory Concentration.

In conclusion, N. arvensis is able to inhibit selectively the expression of the pro-inflammatory chemokine IL-8. The isolation of the active principles purified from N. arvensis could be usefl in identifying safe and innovative pharmaceutical molecules to control lung inflammation in the lung of CF patients. A virtual screening of a furocoumarin (consisting of a furan ring fused with coumarin) compound database has been utilized to select molecules for their ability to inhibit the binding of NF-kB with the consensus sequence identified in the promoter of IL-8. Natural and synthetic organic compounds available in a structure searchable database were retrieved using a substructure searching (2D) method. The distributors accessible on the network that we used are the followings: Sigma-Aldrich, Maybridge, Otava, TimTec, Chembridge and DrugBank. The customized library of 1710 compounds contained 54 structures of angelicin derivatives (MW range: 186-478), 1384 of psolaren derivatives (molecular weight range: 186-619), 51 of Furo (3,2c) chromen-4one structures (MW range: 268-439) and 166 of benzoquinolizin-5-one analogues (MW range: 258-499).

VS approach was followed against both NF-kB p50 monomer and homodimer. The selection of interesting compounds was based on the docking XP GlideScore rank for each target.

Among the compounds exhibiting the highest docking ranks in the dimeric aggregation of the protein, compounds 6f, 7f, 9f and 10f inhibited the formation of a stable NF-kB p50/DNA complex in EMSA studies. The most active compound 7f was a furo(3,2-c)chromen-4one derivative and had an IC50 of about 40 microM. The angelicin derivative 6f and the psolaren derivatives 9f and 10f showed an activity of about 70-100 microM. To determine effects on IL-8 gene expression, IB3-1 cells were pre-treated with compounds 6f, 7f, 9f and 10f 20, 4 and 2 h before infection, at time of infection and 2 h after infection. Total RNA was extracted and the neutrophil chemokine IL-8 transcript was quantified. Compound 7f significantly inhibited the PAO1-dependent transcription of IL-8 in IB3-1 cells at all the time tested. Lower inhibition was observed with the other compounds 6f, 9f, and 10f. No significant modification of P. aeruginosa-dependent IL-8 mRNA induction was observed with compound 8f with doses ranging from 0.1 nM to 100 microM. No changes in cell proliferation have been observed with the compounds 7f, 9f and 10f also at high concentration (100 microM), whereas a significant inhibition was observed with compound 6f at 100 microM.

In conclusion, NF-kB virtual screening of a furocoumarin database was found to be useful identifying novel pharmaceutical molecules able to inhibit the expression of the pro-inflammatory chemokine IL-8, which could be beneficial in controling lung inflammation in the lung of CF patients. Background: Reduction of lung inflammation is an important target of cystic fibrosis (CF) therapy. Glucocorticoids (GC) are anti-inflammatory molecules commonly used to treat inflammation but with controversial efficiency among studies. Upon ligand binding, the glucocorticoid receptor (GR) inhibits (transrepression) or stimulates (transactivation) gene transcription.

The present study focuses on transrepression of activator protein (AP)-1, nuclear factor (NF)-κB activities, transactivation of tyrosine aminotransferase (TAT)3 activity, and interleukin (IL)-8 secretion. IL-8 is a marker of pulmonary inflammation in CF and also a transcriptional target of AP-1 and NF-κB. Thus, measuring levels of secreted IL-8 is an important basis to evaluate the anti-inflammatory effects of GC in CF.

Methods: CF and non-CF bronchial epithelial cell lines were incubated with IL-1β (10 ng/ml) at 4h, 8h, and 16h with or without dexamethasone (dex, 1 µM). Rates of secreted IL-8 were assessed by ELISA 16h after treatment. To evaluate GR transrepression and transactivation at 4h and 8h, we transfected cells using plasmids containing NF-κB, AP-1 or TAT3 promoter coupled to luciferase.

Results: In presence of IL-1β, basal secretion of IL-8 was restored by dex at 16h in non-CF cells, whereas CF cells barely responded. Although other results showed a decrease of NF-κB activity by dex only in non-CF cells at 8h, decrease of AP-1 activity by dex occured only in CF cells at 4h.

Conclusions: Down-regulation of IL-8 secretion in pro-inflammatory context by dex seems to be mediated through either NF-κB or MAPK pathway depending on CF context. Supported by the French Cystic Fibrosis association Vaincre La Mucoviscidose.

Mitran, S.M. Mathematics, University of North Carolina, Chapel Hill, NC, USA A detailed mathematical model of mucociliary clearance has been developed that links cilia motion produced by dynein molecules to overall airway surface liquid (ASL) transport. The model contains a numerical simulation of the three-dimensional motion of thousands of cilia, the viscoelastic ASL velocity fields and the geometry of a portion of airway containing 3-5 branching points. Numerical validation of the model has already been accomplished and the cooperative behavior of cilia that leads to metachronal waves has been reproduced (Mitran, 2007) .

The model is now applied to elucidate the mechanisms of inhaled saline and percussion treatments that are commonly applied to CF patients. Inhaled hypertonic saline treatment is thought to influence the viscosity of the ASL by drawing out water from the epithelial cells through osmosis. This process is simulated numerically by imposing a water flux at the epithelial-ASL boundary and modifying the properties of the ASL in accordance to local concentrations of mucins. Spatial and temporal inhomogeneities in the ASL composition are tracked. The motion of cilia in layers of differing heights and viscoelastic properties is included in the model. A complex interplay between the motor force of cilia, airflow induced stress and ASL properties is observed and the balance of these factors is of importance to clinical CF researchers in understanding the efficacy of hypertonic saline treatment.

The numerical model is also used to model percussive therapy. Chest percussion is captured by repeated modification of the airway geometry over many tidal breathing cycles. The mechanical effects of the percussion lead to local increases in stress at points along the airway. Under normal physiological conditions, these have little effect. It is shown however that for conditions close to stagnation, the additional stress is helpful in re-establishing flow channels of the ASL along the epithelial surface.

cleotidases, NTPDase1 and NTPDase3, are expressed in the airways and profoundly affected by the CF disease.

Methods: Functional evidence for their expression was obtained by enzymatic assays conducted with selective inhibitors on polarized primary cultures of human bronchial epithelial (HBE) cells. Chronic infection and inflammation was reproduced by exposing HBE cultures to supernatant from mucopurulent material (SMM) collected from the airways of CF patients. We compared the responses of NTPDase1 and NTPDase3 to CF and SMM in terms of surface activity, polarity, protein (Western blots) and mRNA (RNAse protection assays and Taqman real-time PCR) expression. These findings were compared to their in vivo distribution in normal and CF airway tissue established by immunohistochemistry.

Results: 1. Experiments conducted on HBE cultures indicated that they account for > 60% of the total ectonucleotidase activity on normal and CF epithelia. NTPDase1 was restricted to the apical surface, while NTPDase3 exhibited a bilateral distribution. 2. Chronic SMM, but not CF, shifted the functional polarity of both enzymes, NTPDase1 being relocated to the basolateral surface and NTPDase3 to the apical surface. 3. In contrast, CF reduced NTPDase1 activity, protein and mRNA levels by 50-70%, while those parameters were raised > 3-fold for NTPDase3. 4. Chronic SMM on CF cultures caused a similar polarity shift and amplified the effects of the disease on their activity and mRNA expression. 5. Their in vivo epithelial polarity in normal airway tissue represented an intermediate between aseptic and SMM-exposed HBE cultures. NTPDase1 was also highly expressed in fibroblasts, as well as endothelial, alveolar and inflammatory cells, whereas NTPDase3 was detected in submucosal glands. 6. In CF airway tissue, their epithelial polarity was shifted in agreement with the SMMexposed HBE cultures, the extent correlating positively with the local incidence of airway plugging.

Conclusion: This study suggests that NTPDase1 and NTPDase3 participate in different aspects of airway defenses against infection according to their biochemical properties. High-affinity NTPDase1 may be recruited to basolateral epithelial surfaces to regulate leukocyte recruitment, while lowaffinity NTPDase3 could reduce the damaging effects of excess ATP release from bacteria and inflammatory cells in the lumen. The down-regulation of NTPDase1 in CF may contribute to the unmanageable neutrophil infiltration characteristic of the disease, which plays a critical role in the irreversible loss of lung function.

Supported by the CFF (PICHER07I0) and the NIH (PPG-P01-HL034332).

Kelley, T.; Shank, S.; Corey, D.; Ziady, A.G. Pediatrics, Case Western Reserve University, Cleveland, OH, USA Nrf2 is a transcription factor that regulates the antioxidant response element and controls redox mediated inflammatory signaling. It is sequestered in the cytoplasm and translocates to the nucleus in response to increases in oxidants. In the nucleus, Nrf2 associates with a number of proteins that promote or diminish DNA binding and transcription. Previously, we investigated the role of Nrf2 on inflammation in a number of cell and animal CF models. Proteomics revealed decreases in antioxidant proteins (APs), especially peroxidases that are promoted by Nrf2, including thioredoxin (TRX) 1, peroxiredoxin (PRDX) 1, 3, 5, and 6, glutathione S-transferase (GST) 1 and 2, and catalase in CF cells and animals vs normal controls by >2 fold. Intracellular H 2 O 2 , a potent inducer of inflammatory cytokine production, was elevated by as much as 5 fold (when stimulated with TNFα/IL-1β) in CF vs normal. Despite increases in oxidants, Nrf2 activation is diminished in CF model cells as evidenced by proteomics and reporter gene analysis. This response is paradoxical, and is responsible for excessive inflammation. Therefore, we studied the activation of Nrf2 in CF. Increases in oxidants are expected to increase Nrf2 nuclear translocation, which is expected to occur normally in CF. However, once in the nucleus, Nrf2 must associate with a number of proteins to facilitate DNA transcription. One of these proteins is CREB binding protein (CBP). Reduced Nrf2 interaction with CPB diminishes its transcriptional activity by increasing its export from the nucleus. Based on previous studies that indicate an upregulation of cAMP signaling in CF cells, we hypothesized that increased phospho CREB (mediated by cAMP) binds CBP and prevents its interaction with Nrf2. We used the cAMP antagonist Rp-cAMPs, which competes with cAMP and blocks activation of its targets. First, we tested the effect of Rp-cAMPs on the expression of proteins regulated by Nrf2 using proteomic analysis. In 3 experiments, Rp-cAMPs significantly corrected (increased) the expression of TRX 1, PRDX 1, 3, 5, and 6, and GST 1 and 2 in the 9HTEo-pCEPR (CF) cells by an average of 4.1+0.8 fold towards normal. Rp-cAMPs decreased H 2 O 2 in 9HTEo-pCEPR cells by 3 fold to a level below that observed in normal cell pairs. This is consistent with an increase in Nrf2 activity and an increase in associated peroxidases. To examine the mechanism by which Rp-cAMPs activates Nrf2, we used Western blot analysis to assay CBP interaction with Nrf2. When we immunoprecipitated CBP and probed for Nrf2 in the 9HTEo-model cell pairs, we found that CBP interaction with Nrf2 was decreased by 62+5.2% in CF. This result was reproduced in nasal epithelia from ∆F508 mice. Treatment of the CF model pCEPR cells with Rp-cAMPs reversed decreased CBP interaction by ~2.5 fold to levels only 23+6.2% below normal. We conclude, that in CF, upregulation of cAMP signaling diminishes CBP interaction with Nrf2, thereby decreasing its transcriptional activity in the nucleus. This phenomenon contributes to the paradoxical dysfunction of Nrf2, reducing antioxidant responses, and allowing for the accumulation of oxidants that promote inflammation in CF models.

Supported by the Cystic Fibrosis Foundation. Background: Cystic fibrosis (CF) is the most common life limiting recessive disease in the U.S., and is due to mutations in the CFTR gene. CF mutations, of which the most common is ∆F508-CFTR, cause massive proinflammatory phenotype in the lung, which is characterized by high levels of interleukin-8 (IL-8). IL-8 is the most powerful known biological attractant of neutrophils, which are called upon to mount a vigorous defense against bacterial invaders. However, even in the absence of bacterial invaders, as occurs, for example, in the lungs of newborn infants with CF, the sustained and spontaneous release of high levels of IL-8 and other related agents causes profound inflammatory damage. The mechanism by which IL-8 gene expression is dysregulated in CF is not known. Our data indicate that in CF, high levels of IL-8 protein are due in part to aberrantly low rates of degradation for the IL-8 mRNA. This mechanism involves the adenine and uridine rich sequences (AREs) in the 3' untranslated region (UTR) of the IL-8 message, which interact with specific ARE-binding proteins. These ARE-binding proteins modulate post-transcriptional processes including mRNA stability and translation of the message into functional protein. Our data suggest that tristetraprolin (TTP), a proto-typical ARE-binding protein, is principally responsible for controlling the stability of IL-8 mRNA. The initial and perhaps the rate limiting step in mRNA decay is the removal of the poly (A) tail from the 3'-end. This is accompanied by removal of the 5'cap structure and is ultimately followed by exonucleolytic (5'-3' and/or 3'-5') degradation of the mRNA body. TTP has been reported to cause enhanced decay of ARE-containing mRNAs by promoting enhanced deadenylation.

Hypothesis: We hypothesize that TTP enhances IL-8 mRNA decay concurrently with multiple cellular signaling pathways.

Method: CF IB3-1 lung epithelial cells and IB3-1 cells stably transfected with plasmid encoding TTP (IB3-1/TTP) were used in this study. To determine the rate of decay of IL-8 mRNA, RNA was isolated from the cells after actinomycin D treatment for various time intervals and the remaining mRNA was analyzed by quantitative real time PCR. The level of β-actin was used as an endogenous control. The rate of IL-8 mRNA decay was studied following the inhibition of MAPK signaling pathway.

Results: TTP protein is expressed in very low levels in CF lung epithelial cells. The transient expression of TTP enhances the rate of decay of IL-8 mRNA resulting in a substantial decrease in the level of IL-8 protein secretion. Moreover, TTP promotes increased destabilization by inducing enhanced deadenylation-dependent degradation. Additionally, IL-8 gene expression in CF cells is also regulated by the MAPK signaling pathways. We found that inhibition of p38MAPK, ERK1/2 as well as PI3K pathways not only reduced the stability of the IL-8 message in CF cells by itself, but also produced an additive increase in TTP-induced IL-8 mRNA destabilization.

Conclusion: Understanding the dysfunctional regulatory mechanism of IL-8 gene expression will lead to more focused anti-inflammatory strategies for CF therapy. Methods: NTHi (10 4 CFU) were inoculated onto the apical surface of 4 different types of air-liquid interface airway epithelial cultures, including human and pig primary epithelia, as well as Calu3 and 16HBE14o-immortalized cell lines. Bacterial persistence, proliferation, and biofilm formation were measured for 10 days following inoculation. Antimicrobial activity of airway surface liquid (ASL) washings was determined by radial diffusion assay. Antimicrobial proteins in ASL were identified using acid urea sodium dodecyl sulfate-polyacrylamide gel electrophoresis (AU-PAGE) overlays followed by mass spectroscopy of microbicidal bands.

Results: Over the 10 days following inoculation in pig primary, Calu3 and 16HBE14o-epithelial cultures, NTHi proliferated by ~1000 fold and formed biofilms. In contrast, bacteria were eliminated in the cultures of human primary epithelia by 24 hours. Only ASL washings from cultures of human primary airway epithelia exhibited antimicrobial activity by radial diffusion assay. AU-PAGE of human primary culture ASL followed by gel overlay of NTHi bacterial lawns revealed a single band with antimicrobial activity. Preliminary analysis of this band by MALDI-TOF revealed histone proteins.

Conclusions: NTHi proliferated and formed robust biofilms in all airway epithelial culture models tested except for human primary airway epithelia. Furthermore, only the ASL of human primary airway epithelia possessed potent antimicrobial activity. Preliminary results suggest that epithelially derived histone fragments in the ASL may play a major role in airway epithelial innate immune defense against NTHi infection. Histones have been shown to play antimicrobial roles in other epithelia. Understanding innate immune factors that prevent NTHi persistence and biofilm formation in the airway may have implications in early CF lung disease. Background: Children with cystic fibrosis (CF) frequently have recurrent lower respiratory tract infections. It is generally accepted that these infections are the result mainly of impaired mucociliary clearance and that immune function in CF may be normal (1) . Although cases of hypogammaglobulinaemia in CF have been reported (2) , the majority of CF centres do not routinely perform serial serum immunoglobulin monitoring. We perform annual measurements of serum immunoglobulin levels in all children attending our tertiary CF clinic. The clinical value of performing immunoglobulins serially is not known. Here we report our data for the last ten years.

Aims: 1. To report the prevalence of hypogammaglobulinaemia in a tertiary CF clinic. 2. To evaluate the clinical benefit of serial serum immunoglobulin measurements in children with CF.

Method: We reviewed the results of serum immunoglobulin measurements over the last ten years in 64 children attending the tertiary CF clinic at the Leicester Royal Infirmary. Patients were classed as having transient hypogammaglobulinaemia if previous low serum immunoglobulin levels returned to the normal range over a period of time without specific intervention. Clinical data was collected for children with low serum immunoglobulin levels.

Results: Thirty boys and 34 girls with a median age of 132.5 months (range 4-192 months) attended our clinic. Sixty-two patients had serum immunoglobulin levels performed annually. Nine of these children had transiently low serum immunoglobulin levels. A two-year-old boy was identified with sustained low immunoglobulin levels. Clinically, he had recurrent severe chest infections and poor growth requiring monthly courses of oral and three monthly courses of intravenous antibiotics. He was commenced on 3-weekly intravenous immunoglobulin replacement therapy (Flebogamma) at the age of 4 years based on his clinical condition and immunoglobulin levels. Over the last three years since initiation of intravenous immunoglobulin replacement therapy, there has been a marked improvement in his clinical condition. He has needed only occasional courses of oral antibiotics and intravenous antibiotics twice. There has also been an improvement in his weight.

Conclusion: The prevalence of persistent hypogammaglobulinaemia in paediatric CF patients at our clinic appears to be low. Transient hypogammaglobulinaemia is relatively common but of uncertain clinical significance. Single measurements of immunoglobulins are therefore of limited value. Serial measurements are necessary to identify the small number of patients with clinically significant hypogammaglobulinaemia. In those children intravenous immunoglobulin replacement therapy may reduce the number of exacerbations of chest infections.

References : Intense inflammation marked by an accumulation of neutrophils in the lung mucosa occurs in patients suffering from cystic fibrosis (CF). Chronic infection with various pathogens including Pseudomonas aeruginosa likely contributes to lung inflammation associated with CF. Neutrophils are recruited to the airspace and breach the mucosal epithelial barrier in an attempt to eradicate P. aeruginosa residing in the airway lumen. Unfortunately, P. aeruginosa is highly resistant to neutrophilic attack and confrontation results in neutrophil release of noxious products culminating in lung tissue destruction. The precise mechanism by which neutrophils exit circulation, travel through multiple tissue compartments, and eventually migrate across the lung epithelial barrier remains to be fully elucidated. Using an in vitro model consisting of multiple cell types, we have previously reported a novel molecular mechanism whereby P. aeruginosa infection on the apical surface of polarized lung epithelial monolayers results in transmigration of neutrophils added to the opposite surface. This process is mediated by epithelial cell polarized secretion of the neutrophil chemo-attractant hepoxilin A 3 . Hepoxilin A 3 is a member of the eicosanoid class of bio-active lipids that is synthesized from arachidonic acid by the actions of a group of enzymes collectively known as 12-lipoxygenases. In the present study our objective was to determine which of the three known human 12-lipoxygenase genes (alox12, alox12B, and alox15) are expressed in lung epithelial cells, and whether expression is modulated by infection with P. aeruginosa. We observed that alox12 and alox15 are expressed under baseline conditions. Both alox12 (4-fold) and alox15 (2-fold) are slightly, but significantly induced upon infection with bacteria for 2 hours. Furthermore, treatment of lung epithelial cells with P. aeruginosa results in a significant increase in epithelial production of the 12-lipoxygenase pathway metabolite 12(S)-HETE. In conclusion, two enzymes capable of hepoxilin A 3 synthesis are expressed in lung epithelial cells and both expression and activity are increased in response to infection. Current efforts are underway to determine which enzyme is responsible for P. aeruginosa stimulated increases in 12-lipoxygenase activity, hepoxilin A 3 production, and neutrophil trans-epithelial migration. Deciphering molecular mechanisms that underlie this complex process may lead to development of more targeted anti-inflammatory therapeutics to alleviate neutrophil associated lung tissue damage.

Zughaier, S.; Tangpricha, V. Medicine, Emory University, Atlanta, GA, USA Background: Vitamin D insufficiency is common among CF patients and may contribute to chronic bacterial colonization of the respiratory tract, leading to chronic inflammation and recurrent infections. Bacterial components like capsular polysaccharides (CPS) and endotoxin (LPS) are shed in large quantities in acutely infected CF lung. These bacterial ligands are potent activators of pattern recognition receptors, mainly Toll-like receptors (TLRs), expressed in host cells. In the normal lung, secreted anti-microbial peptides such as cathelicidin (LL-37) and effective mucociliary clearance serve as the first line of defense for the innate immune system. Controlled levels of pro-inflammatory cytokines are required to alert the innate immune system but excessive release is adversarial to the host. The anti-microbial peptide, LL-37, is found to be significantly higher in induced sputum of CF subjects compared to healthy controls, suggesting impaired killing activity of LL-37 due to inactivation by CF sputum. The aim of this study is to determine the effect of vitamin D3 treatment on macrophage responses in the presence of exogenous LL-37 that mimics the CF lung surface fluid or sputum.

Methods: Human THP-1 cells and murine RAW264.7, and stably transfected epithelial HEK cells with TLR2/6 or TLR4/MD2/CD14 were treated with 1µM of 1,25(OH) 2 D 3 overnight or 1µM ethanol as control, prior to stimulation with TLR ligands LPS and CPS. Cytokine release was measured by ELISA and nitric oxide release was measured by the Griess method.

Results: Upon stimulation with increasing concentrations of TLR ligands, both 1,25(OH) 2 D 3 -treated and ethanol-treated human and murine macrophages produced increasing concentrations of TNFα and nitric oxide release respectively. The addition of exogenous LL-37 resulted in dramatic attenuation of cytokine release in both 1,25(OH) 2 D 3 -and ethanol-treated macrophages. Similar results were seen in epithelial cells HEK/TLR4-MD2 and HEK/TLR2-6. Taken together, the data indicate that LL-37 dampens inflammation measured by TNFα by neutralizing the stimulatory effect of TLR ligands. Treatment with 1,25(OH) 2 D 3 was not additive to LL-37 treatment in the reduction of macrophage derived TNFα.

Summary: The anti-microbial peptide LL-37 decreases macrophage TNFα which may play an important role in the pathophysiology of the CF lung, which demonstrates resistance to LL-37 activity.

Cribb, J. 1 ; Norris, S. 2 ; Moore, P. 2 ; Forest, M.G. 4,1 ; Sheehan, J.K. 3 ; Superfine, R. 2, 1 1. Biomedical Engineering, Chapel Hill, NC, USA; 2. Physics & Atronomy, Chapel Hill, NC, USA; 3. Biochemistry & Biophysics, Chapel Hill, NC, USA; 4. Applied Math, Chapel Hill, NC, USA The constant beating of cilia drives the mucus flow which maintains lung sterility. The interaction between the rapidly moving cilia tip and the viscoelastic mucus layer is largely unknown. However, it is critical for understanding the role of viscoelasticity in mucus clearance and the failure of clearance when mucus becomes too dehydrated. We are beginning to study these phenomena using driven microbead rheology where we place a magnetic micron-diameter bead into polymeric solutions and apply forces to generate dynamics representative of cilia. A spherical bead under constant force in a shear thinning entangled polymer solution experiences a sudden and substantial (>200%) increase in velocity. The earlier and slower steadystate behavior corresponds to known experimental work on transient viscosity overshoots (i.e. strain thickening) in polymer solutions during the startup of shear flow. We use a Stokes Flow continuum approach to model and characterize the instantaneous shear velocity in the field around the bead as it moves through an incompressible liquid. This velocity allows us to calculate the ubiquitous Weissenburg number (Wi) for a field around the bead. We can then determine the dynamic viscosity field and model the expected properties of the surrounding polymer solution. Finally, we consider cilia as a physiological system where this dynamic strain thickening phenomenon may be fundamentally important.

Stick, S.M.; Kicic, A.; Sutanto, E.N. AREST CF, Perth, WA, Australia Rationale: Airway epithelial cells (AECs) play important roles as an effective barrier and in modulating their microenvironment by secreting cytokines & a range of mediators in response to different stimuli. In CF individuals, airway clearance is dysregulated, and together with the persistent inflammation and bacterial infection this eventually leads to pulmonary infection and lung damage. A potentially important stimulant to early airway inflammation is viral infection; with human rhinovirus (HRV) one of the most predominate viral pathogens. This study aimed to evaluate inflammatory responses of CF AECs in response to HRV versus lipopolysaccharides (LPS); & to establish if the responses were CFTR class II mutation-dependent.

Methods: AECs were obtained from CF (all subjects have Class II CFTR mutations, 4.93±1.84 y.o) and healthy (8.37±3.48 y.o) children by non-bronchoscopic brushing. After subsequent passaging, cells were grown in multiwell plates to 80-90% confluent followed by exposure to 10 µg/ml LPS, HRV14 or HRV1B serotypes (MOI 3-25) over a period of 48 hours. Supernatants were collected and IL-8, IL-6 levels measured via ELISA.

Results: CF and healthy AECs produced similar basal expression of IL-8 but CF AECs produced a greater amount of IL-6 than did their healthy counterparts. LPS induced elevated production of IL-8 compared to control cells in both CF and healthy AECs, however there was no significant difference between cohorts. Upon RV1b exposure, CF AECs released markedly higher levels of IL-8 ( Figure 1 ) and IL-6 compared to healthy AEC cultures. Maximal levels of inflammatory cytokines were observed at the highest titer of HRV 48 hours post viral exposure. Exposure of both CF and healthy AECs to increased number of viral particles (MOI) initiated cell lysis, with major cytotoxicity observed in CF epithelial cultures. When compared to HRV1b, HRV14 infection induced a less intense inflammatory response in AECs from both groups.

Conclusions: Results from this study demonstrate that HRV infection induces stronger inflammatory response in CF AECs than LPS stimulation. In response to viral exposure, CF AECs produce significantly higher levels of IL-8 and IL-6 compared to healthy cells. As all of CF individuals partic-ipating in this study are associated with Class II mutation, it may suggest a CFTR-dependent exaggerated IL-8 response to RV exposure. This study also demonstrates the role CF AECs might play in initiating and exacerbating early innate response. Background: Chronic rhinosinusitis (CRS) has a significant impact on the overall quality of life of patients with cystic fibrosis (CF). Treatment strategies vary, but usually include a combination of sinus surgery, systemic and topical antibiotics/steroids, and nasal irrigations. Unfortunately, the indications, timing, and degree of surgical management and postoperative care are controversial. Many CF patients have had multiple sinus surgeries by the time they reach adulthood. The maxillary sinuses, in particular, are a recurrent problem area because normal mucociliary clearance transports against gravity toward the natural ostium. Mucopurulence filling the inferior half of the maxillary cavities is common despite "adequate" maxillary antrostomies. The endoscopic medial maxillectomy (EMM) procedure presented here involves removal of the medial maxillary wall, essentially marsupializing the maxillary sinus into the nasal cavity. This permits physical debridement of mucus in clinic, improved clearance of mucus with nasal irrigations, and increased access for topical delivery of therapeutics. However, surgery is only one part of an overall treatment strategy. The clinical effects of aggressive sinus surgery, proactive clinic debridements, and regimented medical treatment with prolonged intravenous antibiotics and steroids during the healing process have yet to be elucidated. The purpose of the current study was to establish a treatment protocol for severe, recalcitrant CF CRS. Our primary outcome measure was whether patients had improvement in symptoms at an 8 week follow-up interval.

Methods: A prospective cohort study design was initiated to follow results in CF patients treated by this paradigm. Patients meeting Sinus and Allergy Health Partnership criteria for CRS were offered EMM and sinus surgery. Subjects completed a Sinonasal Outcome (SNOT)-22 questionnaire before surgery and at each postoperative visit (2, 4-6, and 8-12 weeks) . Because scarring is facilitated in the presence of extensive infection and inflammation, three weeks of postoperative, culture-directed antibiotics and low dose prednisone 20 mg (4 to 8 weeks) were initiated postoperatively. Nasal saline irrigations were performed twice a day and as needed. Aggressive endoscopic debridements were performed at each postoperative visit.

Results: A total of 11 CF patients (mean age 28 years) with an average of 4.4 prior procedures underwent EMM and sinus surgery for CRS. There were no observed adverse effects or surgical complications. At 8-weeks, there was a significant reduction in overall SNOT-22 symptom scores when compared to preoperative values (62.4 pre vs.36.3 post; p=0.00004). The largest improvements in symptom scores were demonstrated in nasal obstruction (4.1 vs. 0.8), followed by facial pain and pressure (4.3 vs.2.1) and postnasal drip (3.9 vs. 2.1); (p<0.0001).

Early results indicate marked improvement in symptoms with this clinical approach. Ongoing prospective follow-up is necessary to determine whether this treatment paradigm will decrease pulmonary exacerbations in cystic fibrosis. Background: CF lung disease results from a failure of pulmonary innate immune functions. This includes loss of airway antibacterial activity, reduced mucociliary clearance, and failure of immune mechanisms to successfully opsonize and kill bacteria. Cellular inflammation in the CF airway is marked by very high levels of neutrophils. These neutrophils have been attracted into the lung, from the peripheral circulation, by Interleukin-8 (IL-8) which is hypersecreted by CF epithelial cells. We have conjectured that the proteomic and genomic signature of CF neutrophils might generate direct endpoints for both prognosis and therapy. A recent study by Karp and colleagues has identified a protein that is expressed during terminal neutrophil differentiation as an important modifier gene for CF. Surprisingly, Cavaillon and colleagues have shown that airway and peripheral blood neutrophils have similar gene expression patterns. This suggests that the study of neutrophils from the circulation could lead to mechanistic insights into CF pathophysiology. The less intrusive approach of obtaining these cells from the peripheral circulation is attractive from a practical and clinical point of view.

Hypothesis: We have hypothesized that proteomic and genomic differences between CF and non-CF neutrophils contribute to defective neutrophil function in the CF airway.

Methods: CD66+ cells were positively selected from 4.5 -9 mls of whole blood using an immunomagnetic cell separation system (RoboSep platform, StemCell Technologies). Purity of the purified cell populations was demonstrated by FACS and neutrophil morphology was examined using a Wright stain. Protein extracts were prepared in 2D Lysis Buffer and subjected to 2D gel electrophoresis. Proteins were visualized using silver staining. Differentially expressed proteins were determined using the Progenesis software package and identified by MALDI-TOF mass spectrometry. Messenger RNA was isolated and differentially expressed mRNAs determined using the Illumina Human Expression BeadChip.

Results and Conclusions: Use of the RoboSep immunomagnetic cell separation system resulted in the isolation of highly purified neutrophils with 45 minutes of venipuncture. The neutrophils from a subset of the male CF patients displayed "toxic granules" that can form quickly in end stage neutrophils. Several proteins were aberrantly expressed in CF neutrophils including the monocyte neutrophil elastase inhibitor serpin B1, S-100 calcium binding protein A9, leukotriene A4 hydrolase, alpha enolase, desmoplakin, myosin light chain 6 and caspase 14 precursor. Genomic analysis of 24,000 mRNAs clearly distinguished between CF and non-CF neutrophils. The observed proteomic and genomic differences between CF and non-CF neutrophils may provide insight into defective neutrophil function in CF. Nonsense mutations are usually linked with severe disease; however very mild cystic fibrosis (CF) was diagnosed in three siblings homozygous for the mutation E831X, located at the first nucleotide of exon 14a. Now aged 29y (woman), 13y (boy), and 2y (boy), they are pancreatic sufficient, have very mild pulmonary manifestations but have positive sweat tests. Screening of the 27 exons failed to identify another mutation.

RT-PCR performed on mRNA obtained from nasal brushings detect the presence of two mRNA isoforms: a major transcript encoding an exon 14a deleted mRNA and a minor full-length transcript, as confirmed by direct sequencing. The first isoform is predicted to produce a short deletion (from aa 831 to 872) of a cytoplasmic portion spanning between the R domain and the seventh transmembrane segment whereas the second one generates a half-CFTR (truncation after the R domain). Biochemical analysis performed on transiently transfected HeLa cells revealed a truncated protein at the expected size for E831X-CFTR while CFTR del(831-873) generated a coreglycosylated protein. Strikingly, when both mutant proteins were coexpressed, mimicking the endogenous state, a strong fully-glycosylated CFTR band appears at 37∞C without any further treatment. Additionally, co-immunoprecipitation experiments performed with a C-terminal antibody enabled detection of truncated E831X-CFTR. These data suggest an interaction between both proteins inducing a rescuing mechanism of CFTR del(831-873) by E831X-CFTR. The ability of this isoform to correct misprocessed CFTR proteins will be evaluated on F508del-CFTR and other class II mutants. It is still uncertain which isoform bears functional activity accounting for the very mild phenotype observed. E831X-CFTR has been previously described as a functional mutant in Xenopus oocytes (Schreiber et al. Proc Natl Acad Sci USA 1999; 96:5310) however, biochemical data suggest that both isoforms might cooperate and functional studies are still ongoing.

This study underlines the importance of evaluating both mRNA processing and biochemical consequences of a point mutation as E831X mutation generates two distinct proteins. Truncated E831X-CFTR is able to rescue misprocessed CFTR del(831-873), accounting probably for the mild phenotype observed.

This work was supported by public grants from INSERM, EC Grant NEUPROCF (LSHG-CT-2005-512044) Introduction: A common misconception, shared by researchers and the general population alike, is that cystic fibrosis (CF) only affects people of European descent. The diverse ethnic mix in North America suggests that the face of CF might also be changing. We analyzed genome-wide SNP data on 1437 unrelated Canadian CF patients with pancreatic insufficiency (CFPI) from the Canadian Genetic Modifier Study (GC2), shown to be representative of the Canadian CF population, and 1203 deltaF508 homozygous (FF) patients with mild or severe lung function from the US CF Modifier Study using Principal Component Analysis (PCA), which defines subgroups in terms of linear combinations of the SNPs.

Methods: EIGENSTRAT (Price et al. Nat Genet 2006; 38:904 ) was used to perform PCA, and the top 3 significant PCs were examined in 3D plots. We seeded the CF data with SNPs from 6 HapMap3 (http://www.hapmap.org) populations including the Europeans (CEU), East Asians (CHB+JPT), Yoruba (YRI), African Americans (ASW), Indians (GIH) and Mexicans (MEX). We also developed a shrinkage PCA, performed on a shrunken genotype data to reduce the impact of SNPs in high linkage disequilibrium (LD). We applied the shrinkage PCA to the FF samples and compared the results to those from EIGENSTRAT.

Results: No clear structure was observed in the top 3 PCs extracted from the CFPI samples alone. The FF Canadian subset and the FF American sample were distributed randomly in the same space. Seeding with the HapMap3 data, CEU, YRI and CHB+JPT formed three separate tight clusters, while GIH, MEX and ASW formed three short continuous arms pointing to the CEU cluster. CFPI exhibited a tripod structure with most of the subjects forming a cluster at the apex of the tripod, on top of the HapMap3 CEU cluster. This CFPI cluster consisted of mostly self-reported "Caucasian" subjects. However, the rest of the subjects scattered along the CEU-GIH, CEU-MEX and CEU-ASW arms, suggesting population admixture. Furthermore, neither CHB+JPT nor YRI encompassed any of the variation. CFPI and FF samples that distributed along the CEU-ASW arm were either self-reported "African" or "mixed"; and those distributed along the CEU-MEX and CEU-GHI arms were either "Caucasian" or of "Other" ethnicity. On FF samples, the first 2 PCs from the Shrinkage PCA showed a similar tripod structure. Most self-reported African Americans scattered along one arm, while most of the self-reported Hispanics were along another arm. The self-reported Caucasian samples were clustered at the apex of the tripod.

Conclusion: Our PCA suggests the North American CF population contains many individuals with mixed ancestries. Two very different analytical approaches produced very similar results. Given the ethnic complexity of this population, it is crucial to guard against potential spurious association in CF genetic modifier studies, by incorporating either detailed ethnicity information or PCs. PCs present a more practical solution when genome wide data is available, since a full description of mixed ethnicity is rarely available.

Supported by Genome Canada, CF Foundation,Canadian CF Foundation, R01HL068890, R01DK066368. Presented on behalf of the North American CF Gene Modifier Consortium. The search for genetic modifiers of cystic fibrosis (CF) lung disease requires a phenotype measure that accurately categorizes disease severity across all ages, and ideally across different study populations and ascertainment schemes. Forced expiratory volume in 1 second as a percent of predicted based on height, age, and sex (FEV1pp) is a clinically useful and relevant measure of lung disease severity relative to a "normal" or background population. However, the same FEVpp value indicates different levels of CF severity at different ages (eg FEVpp=50 in a 10 year old signals severe disease, while in a 40 year old it is quite mild). Age-specific percentiles of FEV1 among CF patients have also been developed (Kulich et al. Am J Respir Crit Care Med 2005;172:885) which indicate CF lung disease severity relative to other CF patients. The 50th percentile for a 10 year old or a 40 year old indicates average lung function compared to other surviving CF patients, but clearly the 10 year old has more severe lung disease. We propose a new lung phenotype that accounts for the progressive decline of lung function as well as the increasing mortality attrition with increasing age. Using information obtained from the Canadian CF Patient Data Registry (CPDR) we determined survival probabilities to Dec 31, 2002 for each birth year cohort from 1973 to 2002, i.e. from 0 to 30 years of age. For ages over 30 years a logarithmic curve was fit through birth cohort survival estimates based on Canadian live births and CF patients reported alive in the 2002 CPDR. The CPDR survival estimates matched closely with estimates from the US CF Registry which were available up to age 22 years. To calculate the new phenotype for an individual patient, the mean value of available CF age-specific FEV1 percentile scores over a 3 year period was adjusted by the survival percentile corresponding to the mean age of measurements for that patient. This produced a cohort percentile for each patient relative to others who had lower lung function or had died. The new phenotype was computed for 1437 CF patients with pancreatic insufficiency (PI) in the Canadian CF Modifier Study, 1162 deltaF508 homozygous patients with mild or severe lung function in the US CF(UNC/CWRU) Modifier Study, and 1179 CF PI patients in the US CF Sibling and Twin Study. In all 3 groups, the new phenotype was highly correlated with previous phenotypes based on models of longitudinal lung function and predicted age at death (Schluchter et al.) . These complex models require many years of prospective follow-up and produce estimates for an individual that depend on the make-up of the study group included in analyses. The relative ease of computing Kulich CF percentiles and the stable population based survival estimates from the CPDR make the new phenotype easy to implement and most importantly, independent of the study group milieu. Application of the new phenotype in genome wide association analysis will ensure the comparability of results in the different modifier study designs. Supported by Genome Canada, CF Foundation, Canadian CF Foundation, R01HL068890, R01DK066368. Presented on behalf of the North American CF Gene Modifier Consortium. Objective: Diabetes in cystic fibrosis (a.k.a. CFRD) is the most common co-morbidity in cystic fibrosis (CF) resulting in poorer lung function and increased mortality. Genetic modifiers contribute most of the risk of diabetes in CF patients. We tested single nucleotide polymorphisms (SNPs) for association with diabetes in CF using three models (dichotomous trait, diagnosis age using Wilcoxon and log rank statistics).

Methods: As part of the North American CF Modifier Consortium, genotypes for 1796 individuals (children with CF and available parents) in 559 families in the CF Twin and Sibling Study were obtained using the Illumina 610 Quad array. Genotypes for 552,214 autosomal SNPs (96.5%) met stringent quality control criteria. We compared SNP transmission patterns in euglycemic and chronically diabetic CF patients using PBAT. Pre-test power was calculated by excluding children's genotype information. The replication study consisted of 167 CF patients with diabetes and 635 CF controls in the Genetic Modifiers Study.

Results: Without regard to pre-test power, evidence of association was strongest in ZNF98 (P=2 × 10 -7 ; a transcription factor with wide expression including pancreas) and PPAP2B (P=4 × 10 -6 , dichotomous trait; a phosphatase in the Wnt signaling pathway recently associated with type 2 diabetes by pathway analysis). Restricting analysis to SNPs in the top 15% ranked by pre-test power, evidence for association was strongest in CTNNA2 (alpha 2 catenin; P=3 × 10 -6 ; involved in Wnt signaling), VANGL1 (vang-like 1; P=3 × 10 -6 ; involved in Wnt signaling), ATRNL1 (attractin-like 1; P=1 × 10 -5 ; reported to interact with the melanocortin-4 receptor), near PLEKHF2 (P=1 × 10 -5 ), and STX18 (syntaxin 18; P=8 × 10 -5 ). In the replication study, the SNP in CTNNA2 showed evidence of association with the same allele (two-sided P=6 × 10 -3 ). The combined P value (P=2 × 10 -7 ) is statistically significant after a Bonferroni correction for the 82,835 markers of highest power plus markers tested in the replication sample.

Conclusions: A SNP in CTNNA2 shows evidence for association with diabetes in CF. The identification of CTNNA2 along with TCF7L2 (previously found to be associated with diabetes in CF) implicates Wnt signaling in the development of CF-related diabetes. Supported by CFF, NIH, LWPES.

Meconium ileus (MI) is an intestinal obstruction that occurs in about 15 percent of cystic fibrosis (CF) neonates. Analysis of twins and siblings has demonstrated that MI is a heritable trait, indicating that genetic modifiers are largely responsible for this complication of CF. We have previously published the results of a genome-wide linkage analysis which suggested evidence for a modifier of MI at chromosome 8p23. 1 (Blackman et al. Gastroenterology 2006; 131:1030) .

To conduct a regional association analysis, 3,058 SNPs were selected from a 9 Mb region within the previously reported linkage peak where LOD score>1.0. Genotypes were derived from the North American Cystic Fibrosis Modifier Consortium genome wide association study using the Illumina 610 Quad array. Family-based association analysis was performed in 136 families participating in the CF Twin and Sibling Study (CFTSS) in which at least one child had MI. All subjects were pancreatic insufficient (n=269). A cluster of SNPs within the methionine sulfoxide reductase A (MSRA) gene showed association with MI. The highest association was observed for rs11783705, which had an uncorrected p-value of 2.5x10 -5 that nearly achieved region-wide significance. We then tested haplotypes composed of up to 8 consecutive SNPs for association with MI in a 1.85 Mb region containing the MSRA gene. A 3-SNP haplotype (rs10903323 T -rs4840475 G -rs17151637 A; frequency, 0.15) spanning a 3.5 kb region in intron 3 of MSRA was identified as being protective against MI (p=3.1x10 -7 ). In addition, phenotype and genotype data were obtained from the Gene Modifier Study Group (GMSG) for 137 CF patients with MI (110 treated with surgery or enema; 27 as MI on clinical forms) and 117 CF patients verified as not having MI (all subjects were ∆F508 homozygotes). Haplotypes were derived using an estimation-maximization method and case-control analysis showed that homozygosity for the T-G-A haplotype was protective against MI (p=0.02). Odds ratios were constructed to assess the difference in risk between subjects with two copies of the protective haplotype and those with no copies. In the CFTSS and GMSG cohorts, respectively, subjects homozygous for the T-G-A haplotype had 7.2-fold (95% CI, 1.3-72; p=0.016) and 9.5-fold (95% CI, 1.2-427; p=0.035) reduced odds of MI than subjects not carrying this haplotype.

We have identified a haplotype in the MSRA gene that appears to be associated with reduced risk of MI and have replicated this finding in a second CF population. Interestingly, the SNPs comprising the haplotype show little correlation with each other, suggesting that the association is not due to ancestral linkage disequilibrium, but more likely due to a functional variant being harbored on this haplotype.

Supported by the Cystic Fibrosis Foundation and the National Heart, Lung and Blood Institute.

[∆F508]CFTR gene, the disease varies, in a patient specific manner, from mild to severe. It is thought that modifier genes may be responsible, but their identity(ies) remains unknown. Female gender may be a contributing factor. The course of the disease can be distinguished over time by the progressive decline in lung function, defined by "FEV1, % predicted".

Approach: Our approach has been to identify possible genes involved in disease severity by examining the transcriptome of CF lung epithelial cells obtained by bronchial brush biopsy. Modern methods indicate that parallels between changes in mRNA and cognate protein can be as high as 70-80% in eukaryotic yeast systems. The relationship is yet to be fully defined in mammalian cells.

Hypothesis: We hypothesize that hints as to the defective signaling system in the most severe cases of CF can be discerned from analysis of disease-specific mRNAs, and downstream tests of predicted proteins.

Methods: Well-defined CF patients were anesthetized, and bronchial epithelial brush biopsies were obtained by a standard technique. The brushes were placed in TRIAZOL, and the RNA isolated later. Following radiolabeling with 32 [P], samples were analyzed on a cDNA microarray. Results: We have been able to construct a set of statistically validated genomic fingerprints that specifically and significantly discriminate between severe vs. mild CF. Of 1200 informative genes, approximately 80 were found to vary by 2-fold or more as a function of rate of decline of "FEV1,% Predicted". Genes that were found to be elevated in severe vs. mild CF included TGFb1, CCL5 (RANTES), TRADD, IL-8, CASPASE10, a PKA variant, and others. Genes found to be decreased in severe vs. mild CF included histidine decarboxylase (histamine biosynthesis), CCL1, TNFR1, RELA (NFkB), IL-10, ribosomal S6 kinase, caseine kinase 1, and others.

Conclusions: This is the "first ever" transcriptomic analysis of CF lung epithelial cells obtained by acute brush biopsy. These mRNAs have substantial resonance in the CF field, and may serve to identify proteins and pathways that contribute to CF disease severity. To estimate the relative contribution of environmental factors to CF lung disease variation, we compared lung function in affected twins and siblings before and after leaving the parental household.

Methods: Data from 540 pairs of twins and siblings with CF, primarily from the U.S., were collected by the U.S. CF Twin and Sibling Study, and supplemented through the CF Foundation Data Registry (Bethesda, MD). CF-specific FEV1 measures (Kulich et al. Am J Respir Crit Care Med 2005;172:885) were used to derive longitudinal lung function indices for specific living environments (i.e. living with parents vs. living alone). From these measures, intra-familial similarities and differences, and intra-individual differences in lung function were used to quantify the contributions of genetic factors, environmental factors unique to an individual, and environmental factors common to family members. Student's t-test, regression modeling, and heritability estimates were used for analyses.

Results: The heritability estimate, based on 65 monozygous twin (MZ) pairs and 130 same-sex dizygous twin/sibling pairs born within 3 years of each other (DZ/Sib) living together, was 0.57. Furthermore, heritability estimates remained >0.5 over time, confirming genetic influence. Lung function was more highly correlated among MZ twins (0.91) than among the DZ/Sibs (0.51); these correlations did not change after family members moved apart. These intra-familial analyses emphasize the potent genetic contribution, regardless of living environment.

Using intra-familial differences, DZ/Sibs were significantly more different than one another than were MZ twins (together: p = 0.001; apart: p = 0.004), regardless of living environment, again confirming the strength of the genetic component. Regression modeling demonstrated that genetic factors account for 50% of lung function variation, common environmental factors for 14%, and unique environmental factors for 36%.

Using intra-individual differences, regression modeling confirmed the estimates of environmental contributions obtained by intra-familial differences. We also found that the contribution to variation from the common environment is largely derived from living with a sibling with CF (p = 0.006), rather than due to the common environment of the parental household (p =0.532).

Conclusions: Determining sources of variation in lung disease is crucial in decreasing morbidity and mortality for individuals with CF. Through both classic heritability estimates and intra-familial differences, our results confirm the importance of modifier genes, regardless of living environment, and support genome-wide analyses by the CF genome consortium to identify genetic modifiers. As genetic and environmental contributions were roughly equal using intra-familial analysis, our results demonstrate the importance of studying environmental modifiers and ongoing environmental modification to improve outcomes. Although the household environment was found to make less of an impact than unique environmental factors, this may be less applicable to families with a single member with CF. Rationale: Large-scale genomic analyses benefit from a common phenotype for analyses. The North American Cystic Fibrosis Modifier Consortium has developed a consensus pulmonary phenotype based on FEV1 (see Taylor C et al., this supplement), which is cystic fibrosis (CF)-specific and adjusts for survival, age, and sex. Assessing the heritability of this phenotype is important as analyses for genetic modifiers proceed.

Methods: Data from 551 pairs of twins and siblings (1102 individuals) with CF, primarily from the U.S., were collected by the U.S. Cystic Fibrosis Twin and Sibling Study (CFTSS). From these pairs, 67 pairs of monozygous (MZ) twins and 138 pairs of same-sex dizygous twins or siblings (DZ/Sib) born within 3 years of each other were used to estimate heritability using concordance analysis. The pulmonary phenotype was constructed by averaging the 3 most recent years of CF-specific measures of FEV1 (Kulich et al. Am J Respir Crit Care Med 2005;172:885) then adjusting for survival using Canadian CF survival data, and finally transforming the survival adjusted average Kulich FEV1 percentile to a Z-score. Assignments as "A" or "B" within a pair were permuted 106 times to avoid unrecognized assignment bias. Heritability was also estimated from two groups of siblings: (1) in 539 CFTSS siblings in same-sex pairs, dividing the additive trait variance among related siblings by total trait variance for all siblings (as implemented in SOLAR; http://www.sfbr.org/solar), and (2) in a separate collection of 158 sibling pairs from the Canadian Consortium for CF Genetic Studies (CCCGS), by calculating twice the intra-class correlation coefficient (r).

Results: Heritability was estimated by taking twice the difference between the intra-class correlation coefficients (r) in 67 pairs of monozygous (MZ) twins (r = 0.79) and 138 pairs of same-sex dizygous twins or siblings (DZ/Sib) born within 3 years of each other (r = 0.53). The heritability estimate using correlation coefficients was 0.51 [Estimated 95% CI: 0.28, 0.77]. The mean value for the phenotype was not statistically different between MZ vs. DZ/Sib groups via t-test. The heritability estimates generated by siblings alone were 0.80 for the larger CFTSS group and 0.88 in the CCCGS group. The latter estimates are comparable to those obtained from the twin analysis as heritability estimates from siblings are generally inflated as the effect of shared familial environment is not accounted for.

Conclusions: The heritability estimate for the Consortium lung phenotype (0.51) was consistent with previously utilized phenotypes, ranging from 0.54 to 0.82. Heritability estimates for the Consortium pulmonary phenotype demonstrate a strong genetic contribution to lung function variation, which is independent of CFTR genotype.

Supported by funding from the NIH and CFF. A high prevalence of CFTR mutations is observed in patients with diffuse bronchiectasis (DB). Our aim was to evaluate the relation between the presence of CFTR mutations and the degree of nasal transepithelial transport dysfunction in such patients. Nasal potential difference (NPD) was measured in 120 patients with DB of unknown origin, without pancreatic insufficiency and with a normal sweat test ([Cl-] < 60 mmol/L). Extensive CFTR genotype analysis was performed and patients were classified according to the presence of CFTR mutations: none (DB-0), one (DB-1) and 2 mutations (DB-2). In this DB-2 group with normal sweat test and 2 mutations, one mutation at least was classified as mild (class IV or V) or variant, and patients were diagnosed as having cystic fibrosis (CF). A control group comprised 92 classic CF patients with an abnormal sweat test ([Cl-] ≥ 60 mmol/L) and 2 identified CFTR mutations: "severe-CF" patients were homozygous for the F508del mutation and "mild-CF" patients carried at least one "mild" mutation. Results are shown in the table below. Nasal ion transport was normal in DB-0 group. However, in DB-1 group, the Wilschanski's index (which takes into account the different variables measured by NPD) was significantly increased, as compared with the one observed in DB-0 group. Moreover, as the number and/or the severity of CFTR mutations increased from DB-1 to severe-CF, the degree of ion transport dysfunction increased in nasal epithelium and in sweat glands also. These results show a close relation between the severity of CFTR genotype and ion transport dysfunction in DB, both in the nasal epithelium and in sweat glands. One may speculate that the slight abnormal airway ion transport due to the presence of one CFTR mutation may contribute to the development of bronchiectasis if associated with another yet unidentified genetic or environmental factor.

Wilschanskiʼs index : e (response to chloride-free and isoproterenol/ response to amiloride) , a cut off>0.70 predicts CF. ** for comparisons between DB-0 and DB-1: p < 0.01; # for comparisons between DB-1 and DB-2: p < 0.05, ##: p < 0.01; ¶ ¶ for comparisons between DB-0 and DB-2 : p < 0.01, ¶ ¶ ¶: p < 0.0001 (non parametric Kruskall-Wallis test). Cystic fibrosis (CF) is usually diagnosed by a sweat test showing increased chloride concentration. However, new technologies that easily and rapidly show abnormal sweat chloride concentrations are of interest since they would allow performance of the time-consuming and skill-demanding sweat test to be restricted to selected patients. EZscan® is a new patented technology using increasing low direct current voltages (<4V) generating reverse iontophoresis. It allows measurements of electrochemical skin conductance (ESC) which is linked to the ion with the highest physiological concentration in sweat, i.e. chloride. Our aim was to compare this new concept of quick, non invasive ESC measurement to the classical sweat test. EZscan® was performed in 41 adult patients with classical CF (16 F, 25 M; mean age±sd: 28±9 yrs) and in 20 healthy control patients (10 F, 10 M; mean age±sd: 25±10 yrs). ESC was measured by the means of electrodes on each hand and each foot and expressed in µSiemens (µSi). At low voltages (<2V), ESC was analyzed. At the highest voltage (4V), there was a change of ESC that was expressed as dESC, i.e. the difference between the conductance observed at 4V and ESC at low voltages. Sweat stimulation was performed on the forearm by pilocarpine iontophoresis and sweat was then collected on a filter paper. Sweat chloride was measured by colorimetry. Both ESC and dESC were significantly different in CF patients, as compared with control patients: ESC measurements were on the hands: 73±14 µSi and 61±15 µSi (p<0.01), and on the feet: 75±10 and 62±13 µSi (p=0.0001) in CF and control patients, respectively. dESC measurements were on the hands: 9±17 µSi and 49±31 µSi (p<0.0001), and on the feet: 34±21 µSi and 93±24 µSi (p<0.0001) in CF and control patients, respectively. Chloride concentrations as measured by the sweat test were: 95±15 mmol/L and 19±7 mmol/L in CF and control patients, respectively (p<0.0001). There was a significant correlation between dESC and chloride concentration as measured by sweat test (r=-0.78; p<0.0001). dESC diagnostic performance was analyzed by ROC curve modelisation (a cut-off<60 µSi predicts CF). dESC measurement provided a diagnostic specificity of 0.93, a sensitivity of 0.95 and a diagnostic performance as assessed by the area under ROC curve of 0.94. EZscan® appears an attractive tool for easy screening of patients with symptoms evocative of CF: it requires no patient preparation, no medical personnel training; it provides a quick measurement (less than 2 min) and an immediate reading of the results. A study of EZscan® in children will soon be undertaken. 

Data on whether the phenotype of patients with compound heterozygosity for G551D differs from patients with F508del homozygous mutations is divergent. We hypothesised that cystic fibrosis (CF) patients with the G551D mutation would have less severe disease than those F508del homozygotes. We compared the clinical phenotype of patients with a G551D mutation (class 3) with patients homozygous for F508del (class 2) and those with the missense mutation R117H (class 4). Including the R117H mutation allowed a comparison across three mutation groups to be established. Compound heterozygotes for the G551D and R117H were analysed separately. Data were collected for 101 adult CF patients. Group 1-4 represents in order F508del homozygote patients (n=61), those with the G551D mutation (n=13), those with the R117H mutation (n=23) and those compound for both the R117H and G551D mutations (n=4). In the case of categorical variables, binary logistic regression analysis was used to compare group 1 with each of the other three groups and univariate analysis of variance with age as a covariate for the continuous variables. Fisher's Exact test was used to establish any association between genotype and gender. Our findings demonstrate that patients with the G551D mutation and a second severe mutation have a milder clinical phenotype than F508del homozygous patients. Compound heterozygosity for G551D and R117H mutations was associated with normal spirometry, body mass index, no chronic infection and no symptoms. In this context it is possible that mutations on different chromosomes are not independent of each other for the overall impact on phenotype. Interpreting the combination of mutations as opposed to simply accepting the milder mutation as dominant is more appropriate. Objective: In 2006, the American College of Medical Genetics recommended the inclusion of cystic fibrosis (CF) in the Newborn Screening Uniform Core Panel. CF screening in most regions first analyzes immunoreactive trypsinogen and, when elevated, employs a second-tier mutational screen of the CFTR gene. CDC's Newborn Screening Quality Assurance Program (NSQAP) offers proficiency testing (PT) for screening programs that employ CFTR mutation detection. The PT specimens are prepared from CF donor blood. We undertook the validation of CF center genotypes and extensive characterization of a subset of these PT specimens for single nucleotide changes, small insertions and deletions, and large insertions and deletions.

Methods: Specimens were anonymously donated from adolescent or adult CF patients. The promoter, exons, exon/intron borders, portions of all introns, and the 3' end of the gene were sequenced using automated fluorescent DNA sequencing of 49 NSQAP PT specimens to detect single nucleotide changes and small insertions and deletions. Additionally, large insertions and deletions were characterized using a commercially available multiplex ligation-dependent probe amplification assay for all exons.

Results: DNA sequence data was obtained from 98 unique chromosomes from CF patients. In addition to verifying already known mutations, several mutations were identified that were not present in commercially available assays. Four of the PT specimens contained 3 CF associated mutations, however not all 3 mutations were on the ACMG panel. We identified 32 sequence variations with unknown consequences for CF in the 98 chromosomes, four of which have not previously been reported in the CF Mutation database or dbSNP (Intron8:1341+43T/G, Intron12:1899-132G/A and 1899-231T/C, and 3'UTR:4575+94C/T). These 32 sequence variants were used to predict haplotypes from the 98 chromosomes using the two major haplotype blocks defined by the HapMap data (breakpoint of two blocks in exon 10). Interestingly, 29 of the 30 F508del-containing chromosomes showed an identical haplotype across both blocks. The variant F508del chromosome haplotype appeared to be a result of a recombination at the breakpoint of the two blocks.

Conclusion: Newborn screening has been shown to enable prompt detection and intervention in CF patients, resulting in better health outcomes for affected individuals. Extensively characterized PT materials are critical to ensure appropriate PT challenges and the quality of newborn screening laboratory performance. Understanding the haplotype background of CF associated mutations in the U.S. population provides a framework for future phenotype/genotype studies. Objectives: To quantify the role of the host genome to variation in CF respiratory infection.

Methods: Lung infection defined as a single positive respiratory tract culture with any of 13 organisms was determined for 1,354 CF patients in the U.S. CF Twin and Sibling study. Lung infection with Pseudomonas aeruginosa (Pa) or mucoid Pa (MPa) was also established using 4 other criteria: chronic (three positive cultures within six months each one month apart), multiple (3 times within the subject's lifetime), persistent (2 out of 3 cultures positive from cultures obtained in 3 consecutive years) and current (positive culture within three months of the last clinic visit). The contribution of CFTR genotype to infection was assessed in two ways: ∆F508 homozygotes compared to all other genotypes; and ∆F508 heterozygotes with a class 1, 2 or 3 mutation compared to ∆F508 heterozygotes with a class 4 or 5 mutation. The contribution of other genes was derived by concordance and regression analysis of 48 monozygous (MZ) twin pairs and 114 dizygous (DZ) twins or sibling pairs. Measurements and Main results: Odds of infection for 4 organisms (Pa, MPa, Aspergillus (Asp), and Escherichia coli (Ec)) were higher in patients homozygous for ∆F508 compared to other genotypes (OR (95% CI): Pa 2.29 (1.58, 3.32); MPa 1.38 (1.02, 1.88); Asp 1.34 (1.01, 1.78); Ec 1.57 (1.12, 2.19) ). Eight organisms had higher odds of infection in ∆F508 heterozygotes with a class 1, 2 or 3 mutation compared to ∆F508 heterozygotes with a class 4 or 5 mutation (OR (95% CI): Pa: 8.47 (3.77, 19 .02); MPa: 4.33 (2.12, 8.85 ); methicillin resistant Staphylococcus aureus (MRSa): 3.18 (1.46, 6.93); Asp: 4.02 (1.64, 9.84) ; Stenotrophomonas maltophilia (Sm): (4.32 (1.93, 9.71) ; Achromobacter xylosoxidans (Ax): 6.08 (1.85, 19 .98); Ec: 18.06 (2.38, 136.98) ; and Klebsiella pneumonia (Kp): 4.48, 1.34, 15.00)). Concordance rates for Pa or MPa did not differ between MZ twins and DZ twins and siblings for any of the 5 definitions of infection. Of the remaining organisms, only absence of Asp and Kp were more concordant in MZ twins compared to DZ twin/sibling pairs (p = 0.001 and 0.04, respectively). Regression analysis confirmed that MZ twins were less likely than DZ/siblings to have a co-twin infected with Asp (p = 0.046) or Kp (p = 0.041).

Conclusions: CFTR genotype plays a major role in the status of CF respiratory infections by Pa, MPa and numerous other organisms. Similar concordance profiles in MZ twins and DZ twins/siblings for almost all organisms studied indicate that genes other than CFTR contribute minimally to lung infection status in CF.

Gallati, S. 1 ; Ballinari, P. 2 ; Kraemer, R. 1 

Background: A relationship between CFTR genotype and severity and/or progression of pulmonary disease in patients with cystic fibrosis (CF) has proven difficult to establish. This problem may be due to the complexity of the overall estimate of lung function, the diversity of the mutations and genotypes and, as a consequence, the choice of the model of molecular genetic stratification. Characteristics of pulmonary function obtained in CF infants at time of diagnosis have been found to be partially related to the genotype (Pediatr Res 1998; 44 920-926) . In CF children lung function decline is best reflected by pulmonary hyperinflation (Respir Res 2006; 7:138) and increased degree of ventilation inhomogeneities (Am J Respir Crit Care Med 2005; 171:371-8) . A clinically similarly important mechanism is gas trapping, known to impair gas exchange.

Objective: We aimed to determine which physiological factors and lung function parameters (pulmonary hyperinflation, ventilation inhomogeneities, trapped gas, airway resistance, flow limitation) are best associated with genetic prediction models based on molecular genetic criteria such as a) specific genotype, b) nature of mutation, c) functional protein region, d) original classification (classes I-V).

Methods and Findings: Serial annual lung function data obtained from 190 CF patients (95 males; 95 females; ages: 5-18 yrs), provided data representing the degree of pulmonary hyperinflation (FRC pleth ), ventilation inhomogeneities (LCI), trapped gas (V TG ), airway obstruction (sR eff ), and flow limitation (FEV 1 , FEF 50 ). Linear mixed model (LMM) analysis was used to analyze the relationship between each lung function parameter and age, the slope representing progression. Best associations were found between FRC pleth , LCI and V TG and the nature of mutation (inframe, frameshift, nonsense, and missense), as well as between LCI, FEV 1 , FEF 50 and the functional protein region (NBD1, NBD2, MSD1, MSD2 and Rdomain). Taking progression of these discriminative lung function parameters into account, best pulmonary outcome can be predicted for genotypes with at least one missense mutation and both mutations localized in MSD2 and/or NBD2, and worst prognosis was associated with genotypes composed by one inframe mutation in NBD1 (mainly F508del) and one frameshift mutation in NBD2 (mainly 3905ins T). Strongest progression (slope), however, was observed for genotypes including one inframe (mainly F508del) and one nonsense mutation (mainly R553X) both located in NBD1.

Conclusions: The present study demonstrates that extended lung function evaluation is needed to provide evidence for genotype-phenotype associations in CF, and that genotype modeling may be a helpful tool to improve prognostic options even though it is undoubted that many other modifying factors contribute to the CF phenotype. Gene arrays provide a uniquely unbiased survey of the biological state, enabling scientists to identify patterns of change in gene expression that lead to new insights. Though unbiased and potentially powerful, gene array results are inherently ambiguous: in different hands, the same data yield strikingly different interpretations. Broadly speaking, most practitioners identify differentially expressed genes as those with extremely small P values in a statistical test. An alternative practice, recommended by the FDA's microarray quality control project (MAQC), involves selecting genes on the basis of both fold change and P value, using a less restrictive P value threshold. Using four publicly available experiments (1-4) that measured gene expression differences between cystic fibrosis (CF) and non-CF human airway epithelial cells, we compared the two techniques. We found that gene selection on the basis of P value alone generated results that were less repro-ducible than those selected by considering both P value and fold change. Computer simulations demonstrated how consideration of both factors increases concordance. We used the MAQC gene selection approach in an integrated analysis of the four publicly available cystic fibrosis data sets comparing the CF genotype in human airway epithelia to non-CF controls. The analysis did not support the generally accepted hypothesis that the NFkB inflammatory pathway is up-regulated in CF, but revealed that antigen presentation pathway genes, including major histocompatibility complex genes such as HLA-F and HLA-G previously identified by Wright (3), were down-regulated in CF. The identification of this pathway is consistent with clinical observations in CF that the immune response is compromised, and should facilitate research interest in the antigen presentation pathway in CF.

Supported by a CF Research Development Program (STANTO7R0) and the NIH (DK/HL-45881). We are grateful to Dr. Terry Flotte and Chris Mueller for providing gene expression data.

1. Virella-Lowell et al. Effects of CFTR, interleukin-10, and Pseudomonas aeruginosa on gene expression profiles in a CF bronchial epithelial cell Line. Mol Ther (2004) Background: Cystic fibrosis (CF) is one of the best studied human pathologies. However, a comprehensive understanding of CF pathogenesis is still lacking, and the path between the CF-causing mutant ion channel CFTR and the major proinflammatory CF mediator, IL-8, is not fully elucidated. Although the components of CFTR malfunction and CF-inflammation are well characterized, they have failed to generate the knowledge on how they interact resulting in diverse prognosis.

Material and Methods: We employed a Systems Biology approach using GenePattern, Genomatix and Ingenuity Pathways Analysis (IPA), in a study of potential CF biomarkers. CF categorization has been performed using cDNA-microarray derived expression profiles from bronchial brush biopsies of CF patients (n=31), annotated with respect to the rate of decline of lung function (FEV1).

We have used Comparative Marker Selection (GenePattern) to identify highly significant, gender-specific molecular signatures that reflect CF prognosis (False Discovery Rate<0.1, p<0.05). STAT3 is identified by Gene Neighbors (GenePattern) among CFTR-correlated genes indicating that the higher level of mutant-CFTR in the Fast Decline cohort can drive the STAT3-mediated Acute Phase Response (including IL-8) in severe CF. The Fast Decline cohort, in general, is characterized by higher and the Slow Decline cohort -by lower levels of CFTR and IL-8.

The functionality of FEV1-based classification of CF patients has been confirmed with unsupervised Consensus Clustering (GenePattern) with unbiased predictive power (77.4%). This analysis results in three major clusters: Fast, Slow, and Indeterminate Decline. Only 7 out of 31 CF patients are "misclustered" in the "wrong" category that expediently coincides with atypical marker profiles.

We have performed IPA on plausible molecular paths between CFTR and IL-8, and have elucidated a potential role for the estradiol-coregulated NFkB-HSP cross-talk in the CFTR-specific modification of IL-8 transcription. The NFkB and IL-8 associated signaling pathways are distinctly affected in different gender/FEV1-based CF cohorts. NFkB Activation by Virus-es is the most significantly affected pathway, suggesting that perpetuating CF lung inflammation can mimic the NFkB-mediated immune response, normally evoked by pathogens. Finally, we have used Genomatix analysis of IL-8 promoter. Consistent with a CFTR-associated STAT3 profile, and a putative biomarker role for NFAT, the STAT and NFAT binding elements are adjacent to NFKB and, therefore, could compete in IL-8 transcriptosome/enhanceosome function, thus resulting in different levels of IL-8 activation.

In conclusion, this study exemplifies the success of a Systems Biology approach to both CF categorization and to selection of CF biomarkers and modifiers such as TGFB1. Integration of molecular and clinical information in the quantitative framework can provide insights leading to diagnostic/prognostic panels in clinical setting and pathogenetically aimed and personalized treatment in CF. The cystic fibrosis transmembrane conductance regulator (CFTR) gene, which when mutated causes cystic fibrosis (CF), encompasses nearly 200 kb of genomic DNA at chromosome 7q31.2. It is flanked by two genes ASZ1 and CTTNBP2, which have very different expression profiles. CFTR is expressed primarily in specialized epithelial cells, while ASZ1 is transcribed exclusively in the testis and ovary, and CTTNBP2 is highly expressed in the brain, kidney and pancreas, with lower levels of expression in other tissues. Despite its highly regulated pattern of expression, the promoter of the CFTR gene apparently lacks the necessary elements to achieve this. We previously suggested that cis-acting regulatory elements elsewhere in the locus, both flanking the gene and within introns, were required to coordinate regulated, tissue-specific expression of CFTR. We identified a number of crucial elements, including enhancer-blocking insulators flanking the locus, intronic tissue-specific enhancers and also characterized some of the interacting proteins. We recently employed a high-resolution method of mapping DNase Ihypersensitive sites (DHS) using tiled microarrays. DHS are often associated with regulatory elements and use of this technique generated cell-specific profiles of potential regulatory sequences in primary cells and cell lines. We characterized a set of cis-acting elements within the CFTR locus and demonstrated direct physical interaction between them and the CFTR promoter, by chromosome conformation capture (3C). These data provide the first insight into the 3D structure of the active CFTR gene.

Wallace, J. 1,2 ; Stein, Q. 1,2 1. Pediatrics, Sanford Children' Specialty Clinic, Sioux Falls, SD, USA; 2. Sanford School of Medicine, Sioux Falls, SD, USA Introduction: With a frequency of approximately 1 in 14, the M1101K mutation of the CFTR gene is the most common mutation seen in the Hutterites of South Dakota. Delta F508, the only other mutation present in this isolated population, is less common with a carrier frequency of approximately 1 in 72.

Homozygosity for the M1101K mutation is believed to be associated with a classical presentation of the disease often with pancreatic insufficiency although published data regarding phenotype is limited.

Our objective is to present clinical manifestations of the homozygous and compound heterozygous presence of the M1101K mutation.

Methods: Utilizing port CF, individuals either homozygous or heterozygous for the M1101K mutation were identified. Subsequently, review of data from our center's CFF database and individual charts was undertaken to identify the clinical manifestations present in these individuals.

Results: There were a total of 7 individuals homozygous for the M1101K mutation with an age range of 1 to 14 years of age. Six are female and one is male. Age at diagnosis ranged from 3 weeks to 2 years. Two were identified through newborn screening, one following presentation with meconium ileus, 2 due to presentation with failure to thrive and respiratory symptoms and 2 due to family history. Sweat chloride test values have ranged from 89 to 122 mmol. Lung function as measured by FEV1 ranges from 97% of predicted to 126% of predicted. Of these 7 individuals, only one has been hospitalized once for a CF-related diagnosis. Non-mucoid Pseudomonas aeruginosa has been cultured in 4 individuals and methicillin sensitive Staphylococcus aureus (MSSA) has been cultured in all seven.

Five of the seven are pancreatic insufficient and two are pancreatic sufficient. In the 5 older individuals, BMI% ranges from 10 to 84. Impaired glucose metabolism and liver abnormalities have not been identified in any of these individuals.

There are a total of two individuals with the genotype deltaF508/M1101K. The younger one is 8 years of age and was diagnosed at 6 months of age following presentation with failure to thrive and respiratory symptoms. He is pancreatic insufficient and has a BMI of 47%. His FEV1 is currently 102% of predicted and he grows Pseudomonas aeruginosa (nonmucoid and not multi-drug resistant) and MSSA. His sweat chloride test value was 118 mmol. He has had a total of 3 hospitalizations since diagnosis.

The other individual is 46 years of age, was diagnosed based on family history at 12 years of age and had a sweat chloride value of 123 mmol. Her most recent FEV1 is 63% of predicted. She is pancreatic insufficient and has a BMI of 21%. She grows both mucoid Pseudomonas aeruginosa and MSSA on sputum culture and has averaged one hospitalization a year over the past 5 years.

Neither has exhibited abnormalities of glucose metabolism or elevations in liver function tests.

We have presented the clinical manifestations observed in a population of individuals with the M1101K mutation present either in the homozygous or compound heterozygous state. The M1101K mutation shows variable expressivity with the majority of patients having pancreatic insufficiency. (1) . The severity of the phenotype depends on the length of the IVS-8 (polyT) tract which determines the efficiency of splicing. It has been proposed that the R117H mutation should not be included in CFTR panels used in neonatal screening, as it is difficult to predict the outcome for these patients (2) . The R117H mutation is particularly common in N Ireland (NI) so in this study we sought to determine the progression of disease in current patients with the R117H mutation.

Methods: Cystic fibrosis (CF) patients with a F508del/R117H genotype attending the NI Paediatric CF centre from 1983 were identified from the local paediatric CF database, and matched according to age and gender with two homozygous F508del patients. Retrospective annual assessment data (including FEV 1 , pancreatic status, sputum microbiology) was collected between the years of 2003 and 2008.

Results: 446 patients attended the paediatric CF centre from 1983 with 250 of these having either transferred to the adult centre (200) or died (50). The mean age at baseline collection (2003) was 13.2 years (SD 6.9). From this population 5.8% or 26 patients (15 male) were identified as carrying a F508del/R117H mutation. Two patients and their matched pair were not analysed due to insufficient data. There were IVS-8 (poly T) results on 19 patients; 13 were IVS-5T, 6 IVS-7T. R117H patients had lower sweat chlorides at diagnosis (mean chloride 71.6 mmol/l) compared to homozygous F508del (mean 109.3 mmol/l, p<0.001). R117H patients had a significantly higher baseline FEV 1 with a mean of 91.7% predicted compared to 68.3% predicted in homozygous F508del patients(p<0.001), and the rate of decline in FEV 1 showed a significant age/gene interaction, with the F508del patients showing a statistically significant faster rate of decline over time than R117H. Five of 24 R117H patients acquired P. aeruginosa compared with 23/46 F508del patients (p=0.009); and 10/24 R117H were pancreatic insufficient compared with 46/46 F508del (p=0.001).

F508del/R117H patients have significantly better lung function and a slower rate of decline than homozygous F508del patients, but are at risk of developing characteristic CF complications that require treatment and follow-up in a CF centre.

References : The diagnosis of cystic fibrosis (CF) is based on typical clinical features plus evidence of cystic fibrosis transmembrane conductance regulator (CFTR) gene dysfunction (abnormal sweat chloride concentration or nasal potential difference) or identification of two disease causing mutations on both alleles. The diagnosis of non-classic CF is a challenge as CFTR gene dysfunction is not always evident.

Methods: We retrospectively reviewed charts from 1998-2009 of patients who underwent complete gene sequencing of the entire coding region and splice sites of the CFTR gene. Six patients with two abnormalities in the CFTR gene, at least one of which was the 5T allele, were identified. Pancreatic sufficiency was established by fecal elastase (>200 mcg/gm).

Results: Baseline characteristics are presented (table) . The median follow-up time was three years (range 1-11 years). All of our patients except for one were heterozygous for a disease causing mutation and the 5T allele in trans. One patient was homozygous for 5T. TG repeat analysis was available in four patients. Two patients had an elevated immunoreactive trypsinogen (IRT) during newborn screening, and four patients presented with chronic pulmonary symptoms. All patients except one had borderline sweat testing. Two patients (patients 2, 5) underwent bronchoscopy. Bronchoalveolar lavage fluid (BAL) cytology revealed minimal neutrophils and no growth on cultures.

Discussion: Non-classic CF is difficult to diagnose based on initial presentation and testing. This is especially true at a younger age when there is overlap in clinical presentation with other common respiratory diseases, such as asthma. Except for patient 1, none of our patients can definitively be diagnosed with CF or non-classic CF. Approximately five to ten percent of the Caucasian population are carriers of the 5T allele. The expected frequency of individuals with a heterozygous mutation in the CFTR gene and with 5T may be higher than the prevalence of CF, suggesting that other factors may contribute to the disease.

Conclusion: There are no established criteria for the management of non-classic CF. The majority of individuals heterozygous for a CFTR gene mutation and the 5T allele has either no lung disease or develops late onset lung disease. The need for early intervention is unclear, and diagnosis may take years. It is essential that these individuals are monitored regularly at CF centers with sputum cultures, pulmonary function testing, and testing for pancreatic insufficiency. Bronchoscopy with lavage should also be considered for airway cytology (neutrophilic inflammation) and CF associated bacteria.

We aimed to develop an early and non-invasive method for prenatal diagnosis (NI-PND) of Spinal Muscular Atrophy (SMA) and cystic fibrosis (CF) through the genetic analysis of circulating fetal trophoblastic cells (CFTC), thereby avoiding the reference chorionic-villous sampling (CVS) approach which carries a 2% risk of miscarriage.

Methods: We performed phase III clinical validation studies of the ISET (Isolation by Size of Epithelial Trophoblastic cells) strategy for noninvasive prenatal diagnosis of SMA and CF. Individual CFTC were identified from 77 pregnant women including 63 at risk of having a child affected by SMA or CF by using short tandem repeat (STR) genotyping and subsequently used to perform blindly NI-PND of CF and SMA. In parallel, we tested weekly (4th to 12th week of gestation (WG), with ISET and single cell STR genotyping, the blood of 14 women after in vitro fertilization (IVF) in order to assess the kinetics of CFTC circulation in maternal blood.

Findings: A total number of 958 CFTC have been identified by genotyping in the 77 pregnant women. Prenatal diagnosis was performed blindly on 485 CFTC isolated from the 63 tested mothers. Results of NI-PND were demonstrated to be identical to those obtained by the invasive CVS approach. Sensitivity, as computed on the 105 CFTC with a mutated genome, was 100% (95%CI: 94.8-100%). Specificity, as computed on the 380 CFTC without a mutated genome, was also 100% (95%CI: 98.5-100%). The study also identified 473 CFTC in women after IVF. CFTC were found in 4 out of 8 mothers tested at the 4th WG and in all mothers from the 5th WG through the 12th WG.

Interpretation:

These results should open the way to implementation of an early, non-invasive and reliable approach for prenatal diagnosis of SMA and CF. The CFTR (cystic fibrosis transmembrane conductance regulator) gene is a tightly regulated and differentially expressed transcript in many mucosal epithelial cell types. A significant percentage (6%) of cystic fibrosis (CF) alleles remains unidentified even after extensive studies of the CFTR gene by PCR procedure.

Genomic variations in the intronic regions affecting mRNA splicing are increasingly reported. Incorrect splicing can generate aberrant transcripts and also modulate the level of normal CFTR transcript.

In the present study we investigate the molecular defect, by CFTR mRNA analysis, of 11 CF patients, in which only one or no allele were identified by DNA analysis of the whole CFTR gene.

RNA was extracted with TRIzol reagent, from nasal epithelial cells collected using cyto-brush from 11 CF patients and 5 non-CF controls. First strand cDNA was synthesized using hexanucleotide primers and high capacity cDNA Archive kit. The cDNA was amplified in six overlapping fragments spanning the entire gene and then visualized on agarose gel for identifying large deletion/insertion and the product of possible alternative splicing; each fragment was next sequenced and cloned into plasmid vectors.

The mRNA analysis has allowed us to detect the same cryptic exon in two patients (18%): the new pseudo-exon is a 101 bp insertion located between exons 10 and 11 at RNA level. We believe that an identical point mutation identified in intron 10 causes a new donor splice site. The molecular characterization of RNA from nine remaining patients is still ongoing.

Our results confirm the usefulness of RNA analysis to check whether DNA mutations affect normal splicing processes and transcription rates or whether they are solely polymorphisms.

Cuppens, H.; Vliegen, L.; Cassiman, J. Center for Human Genetics, KULeuven, Leuven, Belgium

Next generation sequencing technologies have been recently introduced. However, this technology was initially developed for whole genome sequencing purposes.

We have adopted this technology for complete sequence analysis of the CFTR coding region.

For a 50x coverage, only half a million nucleotides are needed for CFTR sequence analysis of one patient. Therefore, 100-200 patient samples should be pooled in order to use the full capacity of the GS-FLX system. We have developed an economically feasible universal sample tagging approach allowing the pooling of 100 samples with one set of 260 primers (60 amplicon-specific primers and 200 tagging primers). This compares to 6000 primers if amplicon-specific primers are tagged as such.

Normally, each amplicon should be amplified individually and purified. Then, the concentration of each solution is accurately determined in order to pool all amplicons in equimolar concentrations. The universal tagging approach resulted in a less dynamic range of the yield of the different amplicons, so that the time-consuming preparation of equimolar solutions could be omitted.

The time-consuming preparation of individual amplicons could be further simplified by pooling at least 4 DNA samples in PCR reactions, which are then used in a GS-FLX mutation scanning approach. Only amplicons that carry a mutation are then analyzed in a second step in order to determine which DNA sample(s) of the pool carried the mutation(s).

We are further simplifying the PCR steps through the development of robust multiplex PCR reactions. Indeed, 30 amplicons should be analyzed for the CFTR gene, and this can ultimately be only economically feasible if amplified in one, or a limited number, of multiplex PCR reaction(s).

Specifically, we have developed a robust multiplex amplification assay in which biotinylated amplicon-specific primers are locally restricted through streptavidin/biotin crosslinking.

We thus developed an assay using next generation sequencing for routine genetic analysis of the CFTR gene in a diagnostic setting. Cystic fibrosis (CF) is the most frequent autosomal recessive disease in the Caucasian population. So far, over 1500 CFTR gene mutations have been described responsible for classical CF and CFTR-related disorders, most in coding regions and some in introns such as 3849+10kbC>T and 1811+1.6kbA>G. However, a number of cases remain unsolved, making genetic counselling difficult, particularly when the diagnosis is uncertain.

We report the case of a French 25y woman with disseminated bronchiectasis, pancreatic sufficiency and borderline sweat tests. Screening of the 27 exons and search for gene rearrangements only identified F508del in heterozygosity. Investigation at the mRNA level was then performed.

RNA was extracted from nasal epithelial cells, and cDNA was studied in a two steps RT-PCR and amplified in 8 overlapping fragments. By DHPLC in non-denaturing conditions, a longer CFTR mRNA fragment was detected.

Sequencing of this mRNA fragment highlighted a 97bp insert between exons 6b and 7 matching a 101bp sequence of intron 6b with a 4bp (GAAT) deletion. A new stop codon is created within the inserted sequence. Genomic analysis of intron 6b in the patient only revealed the c.1002-1112_1115delGAAT mutation.

In silico studies using bioinformatic tools (Esefinder® 3.0, Human Splicing finder HSF) were used to study splicing signals. This GAAT motif is predicted to be an Intronic Splicing Silencer (ISS) and could be a matrix for the hnRNP A1 ribonucleoprotein. This intronic GAAT cis element helps the spliceosome to ignore pseudo-exons and acts as binding site for proteins promoting exon exclusion. Thus, its suppression could cause abnormal inclusion of a pseudo-exon within the mRNA resulting in CFTR defect.

The amount of abnormal CFTR mRNA in the patient being estimated to be 20% of the wild type by semi-quantitative PCR, the presence of residual normal mRNA could explain the mild CF phenotype. 

Background: Cystic fibrosis (CF) is a genetic disorder caused by mutations in the CFTR gene. The main cause of morbidity and mortality in CF is due to inflammation and chronic infections of the respiratory tract. The CFTR gene is located in the 7q3.1 region and consists of 27 exons that encode the CFTR protein expressed in the apical membrane of airway epithelial cells, pancreas, salivary and sweat glands, intestine and reproductive system, constituting a chlorine channel. There is a low correlation between the type of mutation in the CFTR gene and the clinical features of patients, thus studies of environmental and genetic factors could contribute to a better understanding of the pathophysiology of this disease. The ACE gene encodes the angiotensin-converting enzyme, which is related to the pro-inflammatory response of the lung parenchyma of CF patients. This gene presents a polymorphism denominated D/I, the D allele is characterized by the deletion of 287pb in intron 16 and responsible for a higher gene expression.

Objective: To link the D/I polymorphism of the ACE gene to the clinical markers of CF (Kanga and Shwachman score, body mass index -BMI, age at diagnosis, early pulmonary and gastrointestinal symptoms, microorganisms present, spirometry data, oxygen saturation).

Methods: One hundred thirty-seven patients, 72 (52.6%) males, mean age 15 years, diagnosed as CF by two modified sweat tests. Polymerase Chain Reaction (PCR) technique was used for D/I polymorphism analysis. Statistical analysis was performed using the Statistical Package for Social Sciences (SPSS) version 10.0. The quantitative analysis of non-normal distribution data was performed by Kruskal-Wallis. The quantitative analysis of the normal distribution data was performed by factor variance analysis. Qualitative data were analyzed by the chi-square test, using, when necessary, the Yates correction or Fisher's exact test. In addition the Odds Ratio (OR) and Relative Risk (RR) was calculated. A 5% significance level was used.

Results and discussion: For patients bearing the allele D, the average age for establishing diagnosis was before three years of age (p= 0.04) compared to homozygous II individuals, OR: 3.07 (CI= 1.1 -2.6). The same occurred with the appearance of digestive manifestations (p= 0.05, OR: 8.2, CI= 1.4 -1.46). Regarding the Shwachman score, the main marker of clinical severity in CF, a greater number of patients with the DD genotype classified as severe (p= 0.02), OR: 6.38 (CI= 1.19 -34.21) and RR: 4.81(CI= 1.09 -21.16) was observed. These results suggest that an increased expression of ACE resulting in the DD genotype leads to greater lung damage due to inflammation. We observed that there is a greater number of patients with genotype II infected by the bacteria Achromobacter xylosoxidans (p= 0.03), OR: 4.5 (IC= 1.20 -17.05) and RR: 3.51 (CI= 1.27 -9.74, which leads us to believe that the DD genotype apparently protects patients against chronic infection.

Conclusion: The D/I polymorphism in the ACE gene acts as a modifier in CF.

Marson, F.A. 1 ; Bonadia, L.C. 2 ; Ribeiro, J.D. 1 ; Hessel, G. 1 ; Bertuzzo, C.S. 2 ; Ribeiro, A.F. 1 1. Pediatrics, UNICAMP, Campinas, Brazil; 2. Genetics, UNICAMP, Campinas, Brazil Introduction: Cystic fibrosis CF is a genetic alteration with primarily pulmonary and pancreatic manifestations. The genotype-phenotype correlation in CF has been the objective of many studies, and is most frequent when there is pancreatic insufficiency. There has been no correlation observed between the course and severity of pulmonary expression and CFTR genotype. Pulmonary disease may be influenced by both environmental and genetic factors. Modifier genes may influence the severity of CF phenotype through several mechanisms. For the present study candidate modifying genes related to the mechanism of glutathione action, an important antioxidant in the body, were chosen.

Objective: To link the presence of polymorphisms of both GST genes (M1, T1 and P1-conjugates that promote oxidative stress with glutathione) and GCLC genes (encoding a subunit of the glutamate-cysteine ligase enzyme that limits glutathione synthesis) with the degree CF severity.

Methods: One hundred thirty-seven CF patients, 72 (52.6%) males, mean age: 15 years. Polymerase Chain Reaction (PCR) technique was used to study the GSTM1 and GSTT1 genes, in addition, restriction enzyme was used for the GSTP1( polymorphism 313A/G) and GCLC (polymorphisms 129C/T and 350A/G) genes. Data were correlated with clinical markers [Kanga (EK) and Shwachman scores (ES), Body Mass Index (BMI), age at diagnosis, onset of symptoms(lung and gut), isolated microorganisms, spirometry and saturation oxygen (SaO 2 )]. Statistical analysis: Kruskal-Wallis, Mann-Whitney, ANOVA, T-Test, Chi-square/Fisher's Exact Test. Odds ratio (OR) and relative risk (RR) were calculated when possible. Significance level: 5%.

Results and discussion: For the GSTM1, GCLC (129C/T polymorphism) and GSTP1 gene no statistically significant correlation with clinical markers was observed. For the GSTT1 gene, individuals with at least one allele coding, were classified in the low weight category for BMI (p=0.01, OR:3.1, CI=1.4-7.1, RR:1.4, IC=1.1-1.9). When GSTM1 and T1 genes were analyzed simultaneously, individuals with at least one allele coding were classified in the low weight category for BMI (p=0.05, OR:1.5, CI=1.1-2.7). Patients with both genes (M1 and T1) with null alleles had the lowest SaO 2 (p=0.04), OR:4.3 (CI=1.2-20.1) and the worst ranking for ES (p=0.03), OR: 9.0 (CI =1.5-55.1), RR:5.6 (CI=1.56-19.9) with an association of the genotype for clinical severity. The association of bacterial colonization regarding the null allele for T1 and coding to M1 was related to the presence of mucoid and non-mucoid Pseudomonas aeruginosa (p=0.01, OR:3.1, CI=1.3-7.6, RR:1.7, CI=1.04-2.9, p=0.003, OR:4.8, CI=1.7-13.7, RR:2.8, CI=1.2-6.2, respectively). For the GCLC350 (A/G) polymorphism of the GCLC gene an association to genotype A/A with SaO 2 (p=0.0001, RR:5.8, CI=2.3-14.5) was found and this genotype was associated with a worse rating for EK (p=0.02), lower values of FEF25-75% (p=0.01) and FEV1/FVC (p=0.02), and a greater severity regarding FEV1 (p=0.01), OR: 4.6 (CI:1.3-5.2), RR: 2.2 (CI=1.04-4.8).

Conclusion: GSTM1, GSTT1 and GCLC (350A/G polymorphism) genes act as CF modifiers. CF Center, Clinique St Luc, Bruxelles, Belgium In the ATP binding cassette (ABC) transporters superfamily, MRP (Multidrug Resistance Protein)-1 shares the closest homology with CFTR, which is defective in cystic fibrosis (CF) disease. Functional replacement of CFTR by MRP1 has been previously suggested. We investigated the possible association of MRP1 promoter 5'FR/G-260C polymorphism with severity of CF disease. One hundred thirty non-CF subjects and 234 CF patients homozygous for F508del mutation were genotyped by snapshot: no significantly different allelic frequencies between groups were found. Association of polymorphism with disease severity as assessed by FEV1, BMI, diabetes and Pseudomonas aeruginosa (PA) chronic infection was not statistically significant. The CC genotype tended to be linked to higher rate of chronic colonization by PA as well as with earlier chronic colonization by PA but these trends did not reach statistical significance. Allelic frequencies in 20 CF patients were not in relation with MRP1 mRNA levels and basal/cAMPstimulated anion conductance in nasal epithelial cells. Gene reporter assays were performed in two CF and one isogenic non-CF airway cell lines: no relevance of the 5'FR/G-260C variant was found for the MRP1 promoter transcriptional activity.

Our results did not show any statistically significant impact of MRP1(5'FR/G-260C) polymorphism on disease severity and MRP1 transcription.

Supported by: Mucoviscidose ABCF2, Paris-France. Within the CFTR gene, some sequence variations have been shown to affect pre-mRNA splicing. We are interested in exonic sequence variations capable of affecting exon recognition and inclusion. Such events have already been described for exons 12 and 13, where nucleotide substitutions can disrupt exon splicing enhancer or silencer elements, preventing or favoring the recruitment of splicing factors. These nucleotide variations affect the amount of full length mRNA, thus reducing total CFTR protein expression and modulating the severity of the disease.

Transcripts obtained from nasal cells of healthy volunteers carrying R75Q were amplified by overlapping PCRs. Results indicated a partial exon 3 skipping after separation of the PCR product by DHPLC, confirmed by sequencing of the purified fraction. This result was further analyzed using a minigene splicing reporter construct bearing the c.356G>A sequence variation; concomitantly we studied other sequence variations in the vicinity, namely c.355C>T, c.356G>T and c.352C>T. Transfection of these constructs in three different cell lines (293HEK, HeLa and IB3-1) revealed that exon 3 splicing is affected by c.356G>A, c.356G>T and c.352C>T substitutions.

Bioinformatic analysis of these single-nucleotide variations were performed using several computational splicing analysis tools including ESEfinder3.0 and Human Splicing Finder2.4, which pointed out disruptions in exonic splicing enhancers recognized by specific SR proteins (SFRS1, SFRS6 and 9G8). Accurate quantification of the exon skipping was done by qRT-PCR and allowed us to investigate the effect of modulating these predicted SR proteins levels (by siRNA, plasmid and drugs). Overexpression of SFRS1 protein level restored a normal exon splicing pattern compared to WT implicating SFRS1 in the recognition of exon 3.

Our approach, combining different strategies, i.e. in vitro, in silico and in vivo analysis, enables us to target specific splicing factors to correct CFTR splicing defect. The identification of nucleotide substitutions affecting exon recognition will give a better understanding of CFTR RNA processing and will lead to reconsideration of some neutral variants as diseasecausing, thus improving diagnosis accuracy. This work was supported by public grants from INSERM, EC Grant NEUPROCF (LSHG-CT-2005-512044) The CFTR sequence variant 1811+1.6kbA>G, or more specifically 1811+1634A>G, has been reported as a mutation that causes severe cystic fibrosis (CF) with pancreatic insufficiency (Reboul MP et al. J Med Genet 2002; 39:e73) . This mutation creates a new donor splice site in intron 11 that produces a new exon of 49bp between exons 11 and 12, so-called exon 11b (Chillon M et al. Am J Hum Genet 1995; 56:623) . In this study, we report a novel sequence variant, 1811+1643G>T, located only 9bp downstream from the 1811+1.6kbA>G mutation. We detected this novel mutation in 11 unrelated CF patients from a pool of diagnostic samples analyzed between January 2007 and March 2009 by gene sequence analysis. All 11 patients are of Hispanic descent. One is homozygous for the mutation and the other 10 carry a second known CF mutation (three 3876delA, two Delta F508, and one each D1152H, G1244E, G542X, P205S, 3199del6). There were 587 unrelated Hispanic families in this pool, which makes the frequency of this allele approximately 1% (12/1174) in the Hispanic population. Clinical data from the 11 patients indicate that 1811+1643G>T is a deleterious mutation with varying age of onset depending on the severity of the second mutation. For example, one patient whose second mutation is D1152H, was diagnosed only with infertility (azoospermia) and pneumonia at the age of 43. Similarly to 1811+1.6kbA>G, the G to T substitution at 1811+1643 likely creates a new donor splice site, 4bp further downstream, which would produce a new exon of 53bp. CFTR-cDNA studies are currently underway. The diminishing splicing efficiency that is associated with increasing numbers (>11) of (TG) repeats in the CFTR IVS8-(TG)m(T)n haplotype results in decreasing production of functional CFTR protein. We sought to determine whether the number of (TG) repeats at the (TG)(T)5 locus of hypertrypsinogenemic (HT) infants with genotype ∆F508/5T (∆F508 in trans with only a 5T allele and no other known CFTR mutation) correlated with biochemical (immunoreactivetrypsinogen, IRT) and physiological (sweat chloride, SC) measurements. In addition, we explored the correlation between the IVS8 Poly T length (5T, 7T or 9T), when in trans with ∆F508, and IRT.

Methods: Subjects were identified by the California (CA) 4-step algorithm: 1) IRT (≥62 ng/mL), 2) CA-specific CFTR mutation panel, 3) CFTR full gene analysis utilizing scanning-sequencing technology for specimens with only 1 mutation detected in Step 2, and 4) SC testing and follow-up by cystic fibrosis (CF) Care Centers for infants with 2 or more muta-tions/variants, including 5T. Our analysis focused on infants with any one panel mutation detected during Step 2 and at least one copy of the IVS8 5T allele identified during Step 3. ∆F508 was assumed to be in trans with the 5T allele; for other mutations (CFTRmut), parental DNA testing confirming trans phase was necessary for inclusion in the analysis. The distributions of IRT, initial SC and highest SC were analyzed by (TG) length ((TG)11, (TG)12, (TG)13) among cohorts with genotype ∆F508/5T or CFTRmut/5T. In a separate analysis, we compared IRT among HT infants with genotypes ∆F508 or CFTRmut/5T, ∆F508/7T and ∆F508/9T. All comparisons were tested using the Kruskal-Wallis test.

Results: Of 53 HT infants identified between 07/07 and 04/09, 38 infants had genotype ∆F508/5T, with allele frequencies 66%-(TG)11(T)5, 26%-(TG)12(T)5 and 8%-(TG)13(T)5. Among these infants, IRT trend increased with longer (TG) length (medians (ng/mL): 73, 84, 89 respectively). Although these differences did not reach significance, the number of infants studied with >(TG)11 was small. IRT trend increased with decreasing Poly T tract length (medians (ng/mL) for 9T, 7T, 5T: 68, 74, 77 respectively; p<0.05). Initial SC increased with (TG) tract length (medians (mmol/L): 16, 20, 27 respectively; 11 vs. 12, p=0.08; 12 vs. 13, p<0.05; 11 vs. 13, p<0.01) as did highest SC (medians (mmol/L): 18, 20, 47 respectively; 11 vs. 12, p=0.11; 12 vs. 13, p<0.05; 11 vs. 13, p<0.01). When 15 additional HT CFTRmut/5T infants were combined with the ∆F508/5T group, the results were similar.

Conclusions: Increased SC in association with more (TG) repeats supports the hypothesis that the 5T allele phenotype can be modified by (TG) length and that (TG)12(T)5 and (TG)13(T)5 may act as disease-causing mutations. Increasing IRT with decreasing Poly T length supports the observation that the 5T allele can impact biochemical status. Further study is warranted to determine the role of testing for this locus in the assessment of infants with an unclear diagnosis following, or as part of, CF NBS. Synonymous or "silent" genetic polymorphisms (i.e. exonic alterations that do not change amino acid sequence) can influence function in ATP binding cassette proteins (e.g., p-glycoprotein, Science 315, 525-528). When high usage codons are substituted for others that utilize less prevalent tRNAs, the timing of polypeptide assembly (cotranslational folding) may be altered and confer functional protein defects. In the present study, we initiated a search for synonymous polymorphisms in CFTR that may have clinical significance and/or contribute to CF-related pathogenesis. In particular, we investigated "silent" CFTR polymorphisms in the setting of congenital bilateral absence of the vas deferens (CBAVD). CBAVD has been viewed as a mild CF phenotype, and provides a useful perspective from which to evaluate subtle defects in CFTR that could contribute to clinical abnormalities. In addition, because 70% of CBAVD subjects carry at least one defective CFTR allele, but millions of males worldwide are heterozygous for CFTR defects without developing CBAVD, silent polymorphisms could serve as modifiers that underlie vas deferens involution when only a single (non-synonomous) CFTR defect is detected by conventional genotyping. A large CFTR genotype repository (Ambry Genetics, Aliso Viejo, CA) was used to analyze over 5,000 chromosomes, primarily tested as part of CF carrier screening. A prototypic synonymous mutation (P1290P) was observed in 0.82% of CFTR alleles, and selected for further evaluation from among a larger panel of "silent" allelic alterations. We conducted studies of P1290P by cloning the mutant cDNA (4002 A-G) into expression constructs and completed a comprehensive CFTR analysis that included studies of Band C formation, maturational efficiency, limited proteolysis (to evaluate conformation), surface localization, macropatch channel behavior, and fluorescent dye halide efflux assays. These measures of P1290P were largely comparable to wildtype, with the possible exception of a defect in P1290P gating that is being further investigated. The studies provide a framework from which synonymous CFTR mutations can be tested for their contribution to clinical phenotype. Other "silent" mutations identified in our screen are being analyzed by the approaches and methodologies described here.

Supported by CFF and NIH. The association between the clinical manifestations of cystic fibrosis (CF) and causative cystic fibrosis transmembrane regulator (CFTR) gene mutations is continually undergoing investigation. We present a case study with genetic analysis of a unique family as well as retrospective analysis of clinical laboratory sequencing data in order to enhance the literature with regard to the specific CFTR promoter variant, c.-755C>T (also known as -816C>T). This variant was first described by Bienvenu et. al. in 1993, as documented in the Toronto Sick Kids Cystic Fibrosis Mutation Database. The variant was identified in a 45 year-old male with moderate lung disease, bronchiectasis, and Pseudomonas aeruginosa colonization. No other variant was identified. We describe a 10 year-old African American female who presented with failure to thrive, pancreatic insufficiency, weight loss and vomiting. Sweat chloride levels were measured at 63 and 63 mEq/L for the first attempt and, upon repeat 2 months later, were 48 and 50 mEq/L. Comprehensive CFTR sequence analysis revealed only one variant, c.-755C>T. We also detected this same variant in one half sibling. The sibling was asymptomatic and had sweat chloride levels of 11 and 9 mEq/L. A second half sibling had no variant identified, however the sweat chloride levels were 38 and 42 mEq/L. No other identifiable CFTR mutations have been detected for this family. c.-755C>T has been detected in combination with known diseasecausing mutations as well as in the absence of other mutations. This variant has also been detected in almost 1% of over 10,000 studies (S. Keiles, Ambry Genetics), most of whom did not carry a second disease causing mutation. In addition, three families who did carry a second mutation, deltaF508 identified by newborn screening, had very normal sweat tests of 10-13 mEq/L and presented with no symptoms (S. Keiles, Ambry Genetics). Other researchers identify c.-755C>T as non-disease causing (Cardoso et al. 2004; Romey et al. 1999 ). Based on this information and the frequency of the c.-755C>T variant, it is likely to represent a common disease causing polymorphism and not a deleterious mutation. The cause of the symptoms for this family are most likely not CFTR related. Lung disease is a major cause of morbidity and mortality in CF but its variability among CF patients cannot be attributed to CFTR genotype. Heritability studies indicate that variation in CF lung function measures is influenced by genetic modifiers. Linkage analysis can aid identification of genetic modifiers by detecting genomic regions that are shared in siblings with similar lung function measures.

Methods: Genome-wide linkage analysis for heritable cross-sectional and longitudinal lung function measures was performed using 1151 affected individuals comprising 29 dizygous (DZ) pairs and 610 sibling pairs (560 with cross-sectional measure and 476 with longitudinal measure) from 559 families from the CF Twin and Sibling Study (CFTSS). An additional 166 affected individuals comprising 106 affected sib-pairs (all with cross-sectional measure) from 86 families in the Canadian Consortium for CF Genetic Studies (CCFGS) were analyzed for linkage to the cross-sectional measure. All participants were typed by the Illumina 610 Quad Array as part of the North American CF Modifier Consortium Genome-wide Association Study. Genetic markers for linkage were 2507 autosomal SNPs with average genetic distance of 1.4cM that are included in the Illumina HumanLinkage-12 panel. Variance component methods (both multi-point and two-point) implemented in Merlin and SOLAR and regression methods in Merlin were used to assess linkage. Covariate analysis using lung function measures adjusted for BMI, secondhand smoke, delF508 homozygosity and pancreatic insufficiency was performed to evaluate linkage signals.

Results: For the cross-sectional measure (CFTSS and CCFGS combined), linkage is observed on chromosome 1p21.2 to 1p22.2 (logarithm of odds (LOD) =4.3), chromosome 2q35 to 2q37.1 (LOD=3.0), chromosome 7q32.2 to 7q32.3 about 6 Mb telomeric from CFTR (LOD=2.0) and chromosome 20q13.2 to 20q13.33 (LOD=2.6). For the longitudinal measure (CFTSS only), linkage is observed on chromosome 1p21.2 to 1p22.3 (LOD=2.0), chromosome 5q35.2-5q35.3 (LOD=2.6), chromosome 6q24.2 to 6q25.1 (LOD=2.4) and chromosome 14q23.1 to 14q23.3 (LOD=2.6). All LOD scores reported here are from variance component (multipoint) analysis. Covariate analysis of the CFTSS cohort shows that these linkage signals, especially on chromosome 1, are robust.

Conclusions: Multiple regions of suggestive linkage (LOD scores >2.0) are observed for each lung function measure. A region encompassing 23Mb on chromosome 1p displays significant linkage (LOD score>3.7) to crosssectional measure and suggestive linkage to longitudinal measures of lung function in CF.

Supported by the Cystic Fibrosis Foundation and NHLBI. The link between genotype and phenotype in cystic fibrosis (CF) has been widely examined and varies by clinical trait. To better understand this relationship and to prioritize functional analysis of CFTR mutations, the US CF Foundation has funded an international project to collect genotype and clinical data from CF patients worldwide. Information collected is being assembled into an online database that will be a resource for CF patients, researchers, and clinicians.

Methods: We have collected cross sectional data on 35,312 patients from national registries and large clinical centers in the US and Europe. Clinical data collected included non-identifiable demographics, reason for diagnosis, sweat chloride, lung function, pancreatic status, microbiology, anthropomorphics, and other clinical features when available. These clinical features were chosen as they would allow confirmation of CF diagnosis based on the 2008 Consensus Statement and allow assessment of the spectrum of relevant phenotypes (Farrell et al. J Pediatr 2008; 153:54) .

Results: About 1200 different mutations were present among the more than 70,000 CFTR alleles. Considerable cleaning was required to ensure mutations collected from different databases had universal nomenclature, updated to Human Genome Variation Society standards. Mutations not previously described in the literature or on the CF Mutation Database were selected as possible new variants, or incorrect entries. F508del was the most common mutation, occurring on 43,837 chromosomes for an allele frequency of 62.1%. G542X, G551D, N1303K, W1282X, and R117H all had allele frequencies of greater than 1%. The most frequent 50 mutations account for 94% of alleles, and the most frequent 275 mutations (all seen on 5 or more chromosomes) account for 98% of all alleles. At least 700 mutations are seen only once. As an example of the clinical distribution associated with genotypes, 14,822 patients homozygous for the F508del mutation had a mean sweat chloride of 102.56 mEq/L, 98.1% were pancreatic insufficient, and had an average FEV1% predicted of 74.56. The 196 individuals carrying one copy of 3849+10kbC->T, a common splice mutation, with F508del on the other CFTR gene had a mean sweat chloride of 66.21 mEq/L, 35.1% were pancreatic insufficient, and had a mean FEV1%predicted of 69.12.

Conclusions: We have assembled a cross sectional database of genotype and key phenotypic information for 35,312 patients which will allow comparison across numerous variables. Two hundred seventy-five different mutations constitute the majority of CFTR alleles occurring in CF patients. Rare mutations (occurring in less than five patients) constitute 3% of the total CF patients currently entered in registries. Both these rare mutations and more commonly seen mutations with a variable clinical phenotype will be the focus of further investigation with functional testing to characterize their potential disease liability.

Supported by the US CF Foundation. Among more than 1500 clinically important mutations identified in the cystic fibrosis transmembrane conductance regulator (CFTR), the most prevalent involves deletion of phenylalanine at CFTR position 508 (∆F508). The ∆F508 mutation confers CFTR misfolding, and leads to retention in the endoplasmic reticulum, enhanced degradation, and consequent loss of functional Clchannels in epithelial membranes. Studies to identify small molecules that correct ∆F508 folding defects and enhance surface localization of ∆F508-CFTR have been viewed as priority. However, high throughput screens based on functional assays that measure CFTR dependent Cltransport have often failed to identify consistent "correctors" of ∆F508-CFTR misprocessing due to methodologic limitations. In the present study, we developed a novel, cell-based imaging assay that enables high throughput screens to identify small molecule correctors of ∆F508-CFTR using the Evotec Opera™, an automated confocal microscope system. This assay is designed to semi-quantitatively estimate surface localization of ∆F508-CFTR through an analysis of two-dimensional images in permeabilized cells labeled with antibodies that recognize internal CFTR epitopes. We used the imaging technology and 3G11 rat monoclonal antibody (www.cftrfolding.org) for epitope recognition of nucleotide binding domain 1, and to detect surface localization of ∆F508-CFTR expressed in HeLa cells. The assay established enhanced surface localization of ∆F508-CFTR specifically after low temperature treatment at 27 o C for 48 hours. In addition, treatment with the chemical corrector Corr-4a (10 µM for 24 hours) resulted in much higher ∆F508-CFTR signals (approximately 2.6 fold) at the plasma membrane compared with vehicle control (0.1 % DMSO), whereas rescue of ∆F508-CFTR to the cell surface was negligible following 24 hour treatment with CFTR potentiators genistein (50 µM) or UC CF -152 (10 µM). Enhanced surface localization of ∆F508-CFTR was confirmed by rescue of fully glycosylated ∆F508-CFTR (Band C) as judged by Western blotting and recovery of cAMP-dependent Cltransport after treatment with low temperature or Corr-4a. Recent findings with small molecules such as VX-770 suggest that responses of mutant CFTR in primary airway epithelial cells may predict effectiveness in human subjects. We are therefore optimizing the assay system for analysis of ∆F508-CFTR in polarizing airway cell monolayers. Our results indicate that automated confocal imaging is well suited for multi-well plate configurations, and can quantitatively measure surface localization of ∆F508-CFTR. This method will provide a means for drug discovery of ∆F508-CFTR correctors that ultimately could be useful for the therapy of cystic fibrosis.

Supported by CFF and NIH. Antibody reagents capable of monitoring dynamic changes in the CFTR folded state are not currently available, but would be useful for studies of cystic fibrosis (CF) cellular biology and/or disease mechanism. One of the barriers to generating conformationally selective reagents stems from difficulties encountered by laboratories attempting to synthesize and purify sufficient amounts of full length CFTR for use as immunogen. In order to obtain adequate protein, CFTR with a C-terminal histidine tag was expressed under the regulatory control of the doxycycline inducible promoter. Large scale protein synthesis following induction was achieved in bioreactors with a yield of ~ 2 -8 pg/cell. This capacity enables milligram quantities of protein purification for antibody production (~0.1 -0.3 mg CFTR per billion induced cells). In order to minimize immune tolerance, the purified protein was inoculated into CFTR -/-(knock-out) mice. Lymphocytes from immunized mice were fused to the HGPRT-deficient mouse myeloma line P3X63-Ag8.653 to generate hybridomas. One hundred forty-six mouse hybridoma clonal supernatants were screened for the ability to recognize WT-CFTR by cell-based ELISA under various cell permeabilization conditions, including mild fixation intended to maintain a native CFTR structural configuration in the plasma membrane. We also assessed each monoclonal for the propensity to detect denatured WT-CFTR by Western blotting. Nine clones have been identified that selectively recognize epitopes in the putative "native" but not denatured CFTR configuration. Six hybridomas appear specific for epitopes that are accessible in denatured (but not properly folded) CFTR, and others are positive for both experimental conditions. We plan to characterize CFTR epitope binding sites of all positive hybridomas in the panel. We will also determine the ability of the new reagents to recognize ∆F508-CFTR, with or without rescue by low temperature (or chemical correction). Identification of conformationally sensitive antibodies will provide a critical research tool and help elucidate mechanisms underlying CFTR misfolding. In addition, antibodies that discriminate properly refolded ∆F508-CFTR can be used to optimize and prioritize drug discovery efforts in the future.

Supported by CFF and NIH. ATP in bile is a potent secretogogue, stimulating biliary epithelial cell (BEC) secretion through apical purinergic receptors. While BECs release ATP into bile, the mechanism is unknown. Indirect evidence in other cells suggests vesicular exocytosis contributes to epithelial ATP release. The aim of this study in live BECs was to determine if ATP release occurs via exocytosis of ATP-enriched vesicles.

Methods: Studies were performed in Mz-Cha-1 human BECs, isolated mouse BECs, and polarized rat biliary epithelial monolayers (NRC).

Dynamic live-cell imaging was performed utilizing 3 strategies: i) high sensitivity CCD camera to detect point-source bursts of luminescence (generated by catalysis of the luciferin-luciferase (L-L) reaction by released ATP); ii) confocal microscopy of quinacrine-labeled ATP vesicles; iii) total internal reflection fluorescence (TIRF) microscopy of single exocytic events, in order to image vesicular ATP release from cell populations, single cells, and the submembrane space of a single cell, respectively.

Results: In the presence of L-L, confluent cells were exposed to hypotonicity (33%) to stimulate ATP release, and in response, abrupt pointsource bursts of luminescence were detected. Bursts of released ATP reached a maximum diameter of 200 µm within 2 secs and then decayed with t 1/2 of 5 secs (n=30). In parallel confocal studies of a single cell, quinacrine staining (to localize ATP stores) revealed a distinct population of vesicles ranging in size from 0.4 to >1 µm. In response to hypotonicity (33 %) ATP-enriched vesicles decreased by 61% versus only 8% in control cells exposed to isotonic buffer (n = 10, p<0.001). Visualization of the submembrane space of a single cell during TIRF, revealed ATP vesicles which vacillate in a Brownian manner under basal conditions. In response to hypotonicity (33%), vesicles move toward the membrane and demonstrate a transient increase in fluorescence intensity followed by rapid disappearance consistent with an exocytic event. Increasing the temperature to 35∞C increased the number of exocytic events during basal (0.75 ± 0.1 events/cell/minute)and hypotonic (1.5 ± 0.1 events/cell/minute) conditions.

Conclusion: These novel multi-scale imaging studies demonstrate for the first time the existence of a dynamic ATP-enriched vesicular compartment in biliary epithelial cells which undergo regulated exocytic release upon mechanosensitive stimuli. Understanding the mechanisms involved in biliary epithelial cell ATP release will suggest strategies to increase ATP in bile, thereby augmenting biliary secretion in the treatment of cholestatic liver disorders such as cystic fibrosis.

*Studies supported by the NIH NIDDK and the Cystic Fibrosis Foundation. Abnormalities in the pulmonary circulation are common in conditions associated with hypoxemia, such as cystic fibrosis (CF), and can lead to pulmonary hypertension (PH) and cor pulmonale, which often worsen prognosis. Multiple factors, such as air trapping, hypoxic vasoconstriction, protease/antiprotease imbalance, loss of capillary bed, and endothelial dysfunction, can contribute to the development of PH in CF. However, the molecular determinants of susceptibility to PH remain elusive. In Scnn1b-Tg mice, airway-targeted overexpression of the β subunit of the epithelial sodium channel (βENaC protein, Scnn1b gene) recapitulates many of the pathological features of human CF pulmonary disease, including airway surface dehydration, impaired mucus clearance and chronic neutrophilic inflammation. Our efforts aimed at identifying modifier genes of lung pathology in congenic Scnn1b-Tg mice led us to discover that, in selected strains, airway mucus obstruction was associated with surrogate markers of PH. In congenic C3H/HeN Scnn1b-Tg mice, airway mucus obstruction was associated with poor survival (20% compared to 80% for congenic C57Bl/6N Scnn1b-Tg mice), histological evidence of pulmonary hemorrhage and a high incidence (10/11 at day 5, 8/8 at day 10, 8/8 at 4 weeks, and 10/10 at 8 weeks of age) of blood contamination in bronchoalveolar lavage (BAL). These findings were corroborated by the presence of hemosiderin-laden macrophages -an indicator of previous/ongoing hemorrhage -in the lung parenchyma of surviving mice (3/10 at 10 days, 2/7 at 4 weeks, and 6/11 at 8 weeks of age). Pulmonary hemorrhage was also observed in congenic BALB/cJ Scnn1b-Tg, which also exhibit very low survival, but not in the C57Bl/6N Scnn1b-Tg strain. Intrigued by this novel phenotype, we sought to identify the cause of pulmonary hemorrhage in C3H/HeN Scnn1b-Tg mice and evaluate whether the mice were developing PH, which can be associated with pulmonary hemorrhage in human pulmonary diseases. Strikingly, 9 day-old C3H/HeN Scnn1b-Tg mice already exhibited increased heart/body weight ratio (0.005 for Scnn1b-Tg mice vs. 0.008 for WT littermates, p=0.0064) and right ventricular enlargement (7/8), as assessed histologically and by two-dimensional echocardiographic scan. Importantly, pulmonary hemorrhage and right ventricular enlargement were absent in C57Bl/6N Scnn1b-Tg mice. The inflammatory infiltrate profile was similar between resistant C57Bl/6N and susceptible C3H/HeN strains. These data strongly suggest that C3H/HeN Scnn1b-Tg mice develop PH, whereas C57Bl/6N Scnn1b-Tg mice do not. The availability of congenic strains with either susceptible or resistant phenotypes will allow us to use wide-range genetic approaches to identify heritable factors that determine protection or susceptibility to PH associated with airway mucus obstruction.

Supported by CFF grant to W.K. O'Neal (CFF ONeal07G0). Progressive pulmonary disease is the major cause of cystic fibrosis (CF) associated morbidity and mortality. While progress has been made in patient care, our understanding of key events in the early onset and progression of CF lung disease are lacking, in part because these events occur in infants and young children. The porcine model of CF provides a unique opportunity to investigate the lung phenotype as it evolves. At birth, neonatal CFTR-/-pigs lack pulmonary lesions commonly associated with CF disease including cellular inflammation, submucosal gland hypertrophy, or obstructive lesions. In CFTR+/+ neonatal pigs, the submucosal glands were most evident in the trachea, where mucin immunostaining reveal MUC5AC to be localized mostly in goblet cells of the surface epithelium with rare staining in submucosal glands, and MUC5B detected principally in submucosal glands with some goblet cell staining in the superficial epithelium. CFTR+/and CFTR-/-pigs show similar MUC5AC/5B distributions. Bronchoalveolar lavage of CFTR-/-pigs 6 -12 hrs after birth showed no significant difference from CFTR+/+ pigs in total cell counts or cytokine levels. We performed large scale transcript profiling in lung tissue and airway epithelia using an Affymetrix® porcine genechip. Unsupervised hierarchal clustering of microarray data comparing the mRNA expression profiles of alveolar, bronchus, and tracheal tissue failed to distinguish samples based on genotype. Furthermore, microarray analysis of airway epithelial primary cultures revealed few significant differences in gene expression between the four genotypes studied: CFTR+/+, CFTR-/-, CFTR+/-and CFTR-/∆F508. Analysis of selected candidate genes detected as differentially expressed is ongoing. These data suggest that the lung phenotypes of neonatal CFTR-/and CFTR+/+ pigs are very similar. Furthermore, while there is little evidence of infection or inflammation in the neonatal CF pig lung, a prospective analysis of animals of different ages and disease states may provide mechanistic insights. We hypothesize that cystic fibrosis (CF)-causing CFTR mutations are present in dogs. To test this hypothesis, large numbers of domestic dogs are currently being screened for the presence of CFTR mutations by using temporal temperature gradient gel electrophoresis (TTGE). To increase the probability of detecting CF-causing mutations, the screening population includes animals with disorders that in humans harbor relatively high frequencies of CFTR mutations (i.e., idiopathic bronchiectasis and pancreati-tis). Three different categories of dogs were screened: 167 at-large animals, 147 animals diagnosed with pancreatitis, and 12 animals diagnosed with bronchiectasis. From these animals, one of four CFTR missense mutations (R812W, P1281T, R1456W, and P1464H) were found in each of 27 dogs. The R812W mutation, which does not have a specific human analog, is in a highly conserved region of the CFTR regulatory domain. This mutation site is flanked by several highly-conserved residues that are the sites of CFcausing or congenital bilateral absence of the vas deferens (CBAVD)-causing missense mutations in humans. Although position 1281 is weakly conserved among species, the P1281T mutation is in a conserved region of the second nucleotide-binding domain (NBD2). The canine CFTR R1456W mutation is analogous to the human R1453W mutation, which has been associated with pancreatitis and panbronchiolitis. The P1464H mutation has not been documented in humans and is in a poorly conserved region of the carboxy terminus. Overall missense mutation frequency was 1 in 12.1 dogs, with the most prevalent mutation, R1456W, occurring in 1 in 23.3 dogs. Mutation frequency was approximately twice as great in the pancreatitis and bronchiectasis animals as in the at-large animals. We have cloned the normal canine CFTR as well as all four of the missense CFTR mutations, and we await the results of in vitro expression assays to determine the magnitude of mutation effects on CFTR function. If one of these mutations results in a severe loss of CFTR function, we will screen reproductively competent dogs to identify a carrier for that mutation that can be bred to produce a canine genetic model of CF.

Supported by NIH HL063302 and Cystic Fibrosis Foundation Grant BALLAR07G0. The development of new animal models of cystic fibrosis (CF) has presented tremendous opportunities for understanding the pathogenesis of CF and developing new treatments. We have recently generated a ferret CFTR knockout (KO) model for which newborn animals develop classic signs of CF including pancreatic disease and meconium ileus (MI). Currently, the lung phenotype in this model is unclear. To gain more information on the bioelectric defects seen in CF ferret airways, we have utilized a tracheal xenograft model to evaluate bioelectric properties of CF and wild type littermate tracheas. Tracheas isolated from newborn kits were cannulated with flexible plastic tubing and xenografted subcutaneously into the flanks of nu/nu mice. Grafts were irrigated weekly to remove excess mucous and functional analysis to assess transepithelial potential differences (TEPD) was initiated at 4 weeks post-transplant. Several bioelectric parameters of these airways were assessed including: 1) amiloride-sensitive Na + permeability, 2) cAMP/forskolin-induced Clpermeability, 3) UTP-induced Clpermeability, and 4) the fraction of Clpermeability that is inhibited by GlyH-101. Each of these parameters of ion permeability has been demonstrated to be altered in human CF airway epithelia and/or CF nasal epithelium. Results from these studies demonstrate that cAMP/forskolin treatment (which activates CFTR) induces a change in the TEPD of WT, but not CF, ferret xenografts. This finding is important since CFTR KO mouse tracheal epithelium has alternative non-CFTR Clchannels induced by cAMP agonists and this has been suggested as one reason they do not have a classic CF lung disease phenotype. Furthermore, administration of the CFTR-specific inhibitor GlyH-101 under Clsecretory conditions (low luminal Cl -, cAMP/forskolin) only led to a change in TEPD in WT ferret xenografts-no change was observed in CF ferret xenografts. Interestingly, no significant differences in amiloride sensitive TEPD were observed between WT and CF ferret tracheal xenografts, suggesting that dysregulation of Na + permeability is not a feature of the CF ferret model. It should be noted, however, that CF human bronchial xenografts also do not appear to have significantly elevated amiloride-sensitive TEPDs when compared to non-CF xenografts. Interestingly, a trend toward elevated UTP-dependent changes in TEPDs were seen in CF ferret xenografts, which is similar to what has been reported in the human CF nasal epithelium. In summary, initial characterization of bioelectric properties of the CF ferret airway suggest that this mod-els retains characteristics seen in the human CF airway and hence may be a good model for the progression of CF lung disease. One important goal of translational cystic fibrosis (CF) research is to determine the amount of DF508 CFTR correction necessary to approximate wtCFTR activity. Defining this relationship would help set benchmarks for DF508 CFTR correction. Previous studies have typically examined nonpolarizing systems and have relied on traditional stimuli (eg: cAMP-elevating agents like forskolin or IBMX) to activate wtCFTR and DF508 CFTR in paired expression systems. WtCFTR regulation in vivo is believed to be through Ado stimulation of A2B adenosine receptors, and the gating defect of corrected DF508 CFTR is postulated to be a target of many potentiating molecules. Thus, defining the amount of DF508 CFTR correction necessary to approximate wtCFTR activity would logically include assessment of A2B AR activation of DF508 CFTR in the presence and absence of potentiator molecules. For our studies, DF508 CFTR was corrected by growth of stably transduced DF508 CFTR-expressing CFBE41o-cells at 27 o C for increasing durations. Monolayers were mounted in Ussing chambers, and cells were stimulated with Ado in the presence and absence of P1 (Vrt532). We then compared the Vrt532 + Ado-stimulated Isc in temperature-corrected DF508 CFTR+ cells with that produced by Ado alone in wtCFTR-transduced CFBE41o-cells. These paired cells are stably transduced with lentiviral CFTR cDNA under regulatory control of the CMV promoter. Our results indicate that when Ado-stimulated currents in wtCFTR+ CFBE41o-were used as a denominator (to represent physiologic CFTR regulation), Ado stimulation alone of temperature corrected DF508 CFTR+ CFBE41o-cells produced 1% -5.43% of wt currents following growth of the DF508 CFTR cells at 27 o C for up to 48 hours. In the presence of P1 (10 µM), Ado-stimulated currents in DF508 CFTR+ cells were 13.8%, 13.4%, 61.28%, 77.88%, and 102.18% of the wtCFTR (Ado alone) at 0, 6, 16, 24, and 48 hours (p<0.01 for DF508 CFTR+ cells + Vrt532 compared with the Ado alone at 16, 24, and 48 hours growth at 27 o C). When using common maximal stimuli for both the wtCFTR+ and the DF508 CFTR+ cells [Ado (10 µM), forskolin (10 µM) and genistein (50 µM)], the percent wtCFTR Isc seen in the DF508 CFTR cells was 3.99%, 5.87%, 18.89%, 34.01%, and 56.07% at 0, 6, 16, 24, and 48 hours of low temperature growth. Complementary C-Band assessments will help clarify the relationship between the percent biochemical correction with the percent wtCFTR Isc produced by DF508 CFTR+ CFBE41o-monolayers after increasing growth durations at low temperature. The results indicate that when using Ado-stimulated Isc in wtCFTR+ CFBE41o-cells as a benchmark, the addition of a potentiator (Vrt532) dramatically increased the Ado-simulated currents in temperature-corrected DF508 CFTR+ CFBE41o-cells across increasing durations of low temperature growth. Short low temperature exposure (16 hours) of DF508 CFTR+ CFBE41o-cells was sufficient to restore >60% of the wtCFTR+ response to Ado in the presence of Vrt532. This approach may serve as a useful surrogate for the examination of DF508 CFTR corrective compounds in preparation for clinical trials. Supported by CFFT and the NIH.

Clancy, J.P. 1, 5 ; Gabriel, S. 2, 5 ; Gentzsch, M. 2, 5 ; Donaldson, S. 2, 5 ; Hornick, D. 4, 5 ; Ahrens, R. 4, 5 ; Lymp, J. 3, 5 ; Hocevar-Trnka, J. 3 Rectal tissue should be explored for CFTR outcome measures to detect CFTR modulator activty in early phase clinical trials for a number of reasons, including: a) high level CFTR expression, b) tissue that is not damaged by disease, c) ex vivo studies (ie. flexibility in techniques to detect/quantify CFTR), d) extensive experience (European research community), e) safety and tolerability, and f) established use of Intestinal Current Measurement (ICM) to detect CFTR. A three site GI Outcomes Consortium has developed shared methodology across sites, including common subject preparation; biopsy forceps; buffers, tissue dissection and mounts, Ussing chambers, voltage clamps and EDC software, reagent content and sequence (10uM indomethacin x 20 min, followed by 100µM amiloride, 10µM forskolin/100µM IBMX, 100µM carbachol, and 100µM bumetanide), centralized data review and scoring, and comprehensive SOPs. Based on careful review of run-in studies, a priori criteria have been established to include/exclude data for further analysis, including initial tissue resistance (in non-CF subjects, tissue resistance < 100 ohm.cm 2 predicted the presence of CFTR dependent Clsecretion), amiloride sensitive Isc (in DF508/DF508 CF tissue, the presence of amiloride-sensitive Isc predicted the absence of cAMP-stimulated Isc), and bumetanide inhibited Isc (defining Cltransport in non-CF subjects, and K + secretion in CF subjects without CFTR activity). Eighty to 85% of sequential biopsy specimens met these criteria for further analysis (n=33:40), producing mean cAMP, CCh, and cAMP + CCh-stimulated Isc of 98.26 ohm.cm 2 (SD 66.39), 158.75 ohm.cm 2 (SD 115.68), and 249.91 ohm.cm 2 (SD 167.48) for non-CF subjects across two study centers respectively. These criteria clearly discriminated cAMP-stimulated Isc between non-CF and DF/DF CF cohorts [non-CF = 111.62 ohm.cm 2 (SD 62); CF = -20.45 (SD 23.20) ]. Treatment of non-CF tissues with indomethacin (10 µM x 20 min) led to higher cAMP-, CCh-, and cAMP + CCh-stimulated Isc in non-CF subjects (p< 0.01 for each parameter compared with non-Indo-treated controls). Applying common methodology across two sites led to similar values for CFTR-dependent Clsecretion (mean cAMP, CCh, and cAMP + CCh-stimulated Isc within 10%, p = NS) that is in line with published reports (Hirtz S et al. Gastroenterology 2004; 127:1085) . Data from the third study site is under analysis following installation of common equipment. Our data indicates that ICM can be performed across multiple institutions with complementary data following standardization of procedures. GI outcome measures are attractive for further clinical development. Supported by CFFT. How the loss of CFTR function is affecting cholesterol processing is currently unknown. Previous studies by our laboratory suggest that a cellular feedback response to lost CFTR function increases the cAMP pathway resulting in cholesterol accumulation. The goal of this study is to further elucidate the mechanisms relating cAMP signaling and control of cholesterol transport. Two indicators of chronically increased cAMP signaling in cystic fibrosis (CF) cells is the increased expression of both pCREB and betaarrestin-2 (βarr2). Since βarr2 actively participates in the internalization and sorting of G protein-coupled receptors (GPCRs), like the β2-adrenergic receptor (β2-AR), into the endosomal/lysosomal pathway, we hypothesize that βarr2 is playing a regulatory role in linking the cAMP pathway to cholesterol processing in CF. To test this, 9/HTEo-cells stably overexpressing GFP-tagged βarr2 (βarr2-GFP) and their respective control cells (cont-GFP) were examined by filipin stain to determine intracellular processing of free cholesterol. βarr2-overexpressing cells exhibit a clear perinuclear accumulation of cholesterol similar to that seen in CF cells and also exhibit a colocalization of the βarr2 and cholesterol indicating a shared processing mechanism. Also, similar to CF cells, the intracellular cholesterol accumulation can be alleviated with the cAMP binding antagonist Rp-cAMPS after 72 h as determined by filipin stain. Furthermore, depletion of βarr2 expression with shRNA in CF cells eliminates cholesterol accumulation. These data taken together show that βarr2 overexpression alone is enough to cause CF-like perinuclear cholesterol accumulation which can be reversed by inhibiting cAMP pathway activation. Because of this, the effect of chronic βarr2 expression on the cAMP pathway was examined. The βarr2-GFP cells demonstrate a 2.2-fold increase in pCREB expression compared to cont-GFP cells as determined by Western blot analysis (p < 0.05) indicating that increased βarr2 expression causes increased pCREB expression. To assess the role of CREB activation in the development of CF-like cholesterol accumulation, 9/HTEo-cells stably expressing dominant-negative CREB (kCREB or CREB-133; mutations which either block binding to CRE or phosphorylation of CREB, respectively) and wildtype cells (wt-CREB; 9/HTEo-cells stably expressing human wildtype CREB) were exposed to the cAMP analog 8-Bromo cAMP (8-Br) for 48 h and examined by filipin stain. The CREB-133 and kCREB cells exhibit no apparent alteration in cholesterol processing in the presence of 8-Br whereas untreated wt-CREB cells show perinuclear cholesterol accumulation which in the presence of 8-Br becomes even more pronounced. To ensure the co-localization of βarr2 and accumulated cholesterol is occurring in the lysosomes/late endosomes, βarr2-GFP cells exposed to lysotracker and examined by filipin stain show a clear co-localization of βarr2, lysotracker, and cholesterol indicating that βarr2 and cholesterol are both being sorted into the lysosomal/endosomal pathway. The same localization of cholesterol is observed in CF cells. All these data implicate a regulatory role of βarr2 on the development of the cholesterol phenotype in CF. Research supported by grants from the Cystic Fibrosis Foundation and the NIH/NHLBI. Mucociliary clearance (MCC), an innate host defense mechanism important for keeping the airways cleared of inhaled particles, bacteria, and viruses, is impaired in cystic fibrosis (CF), facilitating mucostasis and chronic bacterial infections. A number of genetically engineered mouse models have recently been designed to modify various components of the mucociliary apparatus. These models not only allow the study of how alterations in MCC compromise lung defense, but have the potential to be useful model systems to test potential therapeutic regimens directed at stimulating mucus clearance. Traditionally, radionuclide particle clearance has been used to quantify mucociliary clearance rates in the airways of humans and large laboratory animal models. Unfortunately, attempts at performing similar measurements in mice and other small rodents have been less successful due to the increased complexity associated with radioaerosol delivery and imaging in these small animals. The objective of these studies was to develop radionuclide particle clearance methodologies for quantifying mucus clearance in anesthetized and conscious mice.

With deposition by aerosolization, the majority of the label is delivered to the alveolar space. Because alveoli are not cleared by mucociliary transport, this deposition pattern creates an enormous background noise, against which the signal from the airways is difficult to detect. In these studies, we developed the use of very small volume (150nl) "focal deposition" within the large airways to ensure only airway deposition of the radiotracer ( 99m Tcsulfur colloid). Once deposited, clearance of this material was followed over time in either anesthetized or conscious mice by traditional scintigraphy using a gamma camera.

Using these techniques, we observed the classical "biphasic" clearance of radiotracer observed in humans; a rapid phase followed by a more pro-longed slower clearance phase. In animals anesthetized with isoflurane (1.5%), the initial rate of clearance was 5.34±0.4 %/min over the first 10 min followed by a sustained rate of 1.21±0.1 %/min. However, in mice killed with an overdose of isoflurane (> 10%), known to inhibit cilia beating, clearance of radionuclide was essentially inhibited (0.13±0.01 %/min), demonstrating clearance of radiotracer is dependent on the mucociliary clearance apparatus. We have also performed studies in restrained conscious animals, where clearance was noticeably stimulated. These mice exhibited an initial clearance of 12.3±1.2 %/min over the first 6 min followed by a sustained rate of 0.9±0.1 %/min thereafter.

We predict that techniques developed in this study will allow us to make in vivo comparisons of mucus transport between normal and genetically modified mice and other small animals to facilitate the understanding of the physiological regulation of mucus clearance. Furthermore, we envision that these tools will be valuable in assessing the impact of therapeutic regimens directed at increasing mucus clearance in a variety of small animal models, prior to initiating human studies.

Supported by Cystic Fibrosis Foundation Therapeutics, Inc. (But-ton07XX0). CF Center, Hadassah University Hospital, Jerusalem, Israel; 3. CF Center, Carmel Medical Center, Haifa, Israel; 4. CF Center, Rambam Medical Center, Haifa, Israel; 5. CF Center, Shaare Zedek Medical Center, Jerusalem, Israel; 6. CF Center, Soroka Medical Center, Beer Sheva, Israel Background: Ex vivo intestinal current measurements (ICM) have been used to study CFTR function in cystic fibrosis (CF). However, the protocol of pharmacological agents applied to the tissue is under debate; in some protocols histamine is not added. The aim of this study was to evaluate the "Rotterdam protocol"(1) which uses histamine, in a group of healthy control non-CF and CF patients, and in a group of patients whose CF diagnosis is uncertain.

Methods: Patients who were suspected for CF were referred to our laboratory for ICM. These patients (mean age: 4±4 years) had a variety of symptoms suggestive of CF including recurrent pneumonia, unexplained bronchiectasis, chronic diarrhea and failure to thrive. ICM was performed using the Rotterdam protocol which involves, among others, secretagogues such as carbachol and histamine which act as chloride channels activators.

Results: In 14 healthy control patients and 10 CF patients the carbachol and histamine results are 16±7 and 11±7µA/cm 2 for the healthy control group and -4±8 and -3±3µA/cm 2 for the CF group, respectively. Sixty-five patients suspected for CF were divided into 2 groups according to their ICM results: 40 patients with normal ICM (carbachol: 20±9µA/cm 2 , histamine: 11±5µA/cm 2 ) and 25 patients with abnormal ICM (carbachol: 3±7µA/cm 2 , histamine: 0±6µA/cm 2 ).

A group of 33 patients out of the 65 mentioned above had borderline sweat chloride levels (40-60 mmol/L). The carbachol and histamine results for this group are 13±10 and 7±8µA/cm 2 , respectively. In 6 out of these 33 patients (18%) the analysis of ICM results was decisive due to the effect of histamine in addition to carbachol.

Conclusion: Histamine is important in contributing to the accuracy of the ICM measurement particularly in non-classic cases. Larger studies are needed to confirm these results. Background: A wide range of clinical trial endpoints is needed to adequately evaluate proposed therapeutic agents in cystic fibrosis (CF). This is especially critical as new potentiators and correctors are being tested. Although Nasal Potential Difference (NPD) measurements may be of benefit, the inter-operator variability is but one factor that may limit its utility in a multicenter study. Our hypothesis is that measurement of CFTR chloride channel function in rectal epithelial cells in vitro may be more reproducible and a more robust outcome measure compared to NPD.

Aims: To evaluate the potential usefulness of electrophysiological studies on rectal mucosa in rectal biopsy samples ex vivo as a surrogate marker for measurement of CFTR function and response to therapies. These results are compared to NPD results on the same subject.

Methods: Intestinal Current Measurement (ICM) experiments were performed according to the Rotterdam protocol (1) using Physiological Instruments Ussing chamber system and University of Iowa voltage clamp.

Results: Carbachol response in healthy control subjects (n=14 subjects. 20 analyzable biopsies) was 17.2 ± 2.7 µA/cm 2 , histamine response was 11.2 ± 2.0 µA/cm 2 , forskolin was 5.1 ± 1.4 µA/cm 2 . This was reversed in CF: carbachol -4.8 µA/cm 2 , histamine -1.6 µA/cm 2 , with no forskolin response. Five controls performed NPD and ICM. The carbachol response was 16.7 ± 4.6 µA/cm 2 , histamine was 9.8 ± 2.6 µA/cm 2 , forskolin was 2.8 ± 0.3 µA/cm 2 , average basal PD was -11 ± 1.6 mV, chloride-free and isoproterenol response was 13 ± 1.2 mV. The carbachol response correlated well with the basal PD (0.7, p=0.08) and with the chloride free and isoproterenol response (0.63,p=0.1), [Spearman test].

Conclusions: Rectal biopsies allow measurement of CFTR mediated chloride transport and may be used as a therapeutic endpoint for studies. They are comparable with NPD. As ICM may be performed on young children and infants, this endpoint may be more practical than NPD. The tissues may also be used for several biochemical parameters linked to CFTR dysfunction such as conversion of CFTR to the mature form, fatty acid metabolism and inflammatory mediators. Supported by CFF. Expression of CFTR by gene transfer offers the potential to correct cystic fibrosis (CF) abnormalities. Adeno-associated virus (AAV) is a promising vector for CF gene transfer but the length of DNA that can be efficiently packaged in AAV vectors is limited. To produce a CFTR cDNA insert that can be packaged in an AAV vector, we shortened the CFTR coding sequence (CFTR∆R) by deleting the nucleotides corresponding to residues 708-759 of the R domain. Our earlier work showed that CFTR∆R retained the Clchannel properties of full length CFTR and corrected the Cltransport defect in differentiated CF airway epithelia in vitro. Here we tested the hypothesis that the shortened CFTR transgene can rescue CFTR function in vivo and improve a clinical phenotype. CFTR-/-mice develop severe intestinal disease that is usually lethal at the time of weaning. Previous studies showed that expressing wild-type CFTR in the intestine with the intestinal fatty acid binding protein (FABP) promoter rescued the intestinal phenotype. Therefore, we generated CFTR-/-mice that contained a transgene with the FABP promoter driving CFTR∆R. Immunocytochemistry revealed expression of CFTR∆R in intestinal villi and crypts of transgenic, but not CFTR-/-mice.

Immunoprecipitation and phosphorylation experiments also confirmed the presence of a shortened CFTR in the intestine of CFTR∆R transgenic, but not CFTR-/-mice. To test for CFTR function, we studied ileal and jejunal segments from CFTR-/-mice transgenic for CFTR∆R and found that they generated cAMP-stimulated short-circuit currents that were inhibited by bumetanide; the currents were equivalent to those in CFTR-/-mice transgenic for wild-type CFTR driven by the FABP promoter. Importantly, the FABP>CFTR∆R transgene prolonged the survival of the CFTR-/-mice (80-100% survival beyond 120 days). Histopathological analysis showed marked improvement in the intestine of CFTR-/-that carried the FABP>CFTR∆R transgene. These data indicate that a shortened CFTR can correct the electrophysiological and pathological intestinal abnormalities in CFTR-/-mice and rescue their survival from intestinal disease. Thus, these results suggest that a CFTR∆R construct may be a valuable part of an AAV vector for CF gene transfer. Our laboratory recently generated an exon-10 deleted cystic fibrosis (CF) ferret model using CFTR gene-targeted fibroblasts and somatic cell nuclear transfer (SCNT) (Sun, et al. J Clin Invest 2008; 118:1578) . At birth, CFTR knockout (KO) kits demonstrate pancreatic lesions and variable degrees of intestinal obstruction characteristic of meconium ileus (MI). Although pharmacologic approaches to treat MI in these CF ferrets may prove useful in enhancing post-natal survival, developing a genetic background that alleviates MI would be extremely useful for dissemination of the model and characterization of a lung phenotype. To this end, we have begun to develop a transgenic CFTR-KO ferret model that expresses fCFTR containing a 3xHA tag in ECL4 (fCFTR-HA) under the direction of a gutrestricted Intestinal Fatty Acid Binding Protein (iFABP) promoter. We show that indeed iFABP protein is only expressed in the gut of ferrets and expression is activated by E28 of the 42 day gestation. The insertion of the HA-tag in fCFTR was included to allow for sensitive assessment of transgene expression and the generation of a useful animal model for studying fCFTR processing and cell biology in vivo. The fCFTR-HA cDNA has been generated and is fully processed and functional like the hCFTR-HA counterpart. The transgenic construct containing the rat iFABP promoter driving the fCFTR-HA cDNA and a floxed PGK promoter-driven Zeocin resistant gene cassette was generated. Cre recombinase-mediated excision of the PGK-Zeocin cassette will be needed to ensure that the PGK promoter does not interfere with restricted iFABP promoter-directed expression. To test whether this CRE-mediated approach of engineering primary ferret fibroblasts is feasible, we generate pools of CRE-reporter fibroblasts transfected with a CMV-LacZ(LoxP)/AP/Neomycin (Z/AP) expression cassette and tested whether virally mediated expression of CRE was capable of efficiently switching gene expression from LacZ to alkaline phosphatase (AP). These studies demonstrated 95% efficiency of CRE-mediated excision. We are currently undergoing experiments to generated iFABP-fCFTR-HA transfected/selected pools of CFTR-KO female ferret fibroblasts that will then be subjected to CRE-mediated excision of the PGK-Zeocin cassette and used as nuclear donors for SCNT cloning of ferrets. Following nuclear transfer, 28 day embryos will be harvested to generate primary fibroblasts and to determine tissue-specificity of transgene expression in the lung, pancreas, and intestine. Rederived fibroblasts from fetuses that demonstrate intestinal-specific expression of fCFTR-HA will then be used as nuclear donors to generate transgenic iFABP-fCFTR-HA CFTR-KO ferrets by SCNT. These efforts may greatly enhance the utility of the CF ferret model by preventing neonatal complications of MI, thus allowing for studies of lung pathogenesis in the model. The development of new animal models for cystic fibrosis (CF) are needed to dissect mechanisms of disease progression and develop therapies. We have recently obtained births from a newly developed CFTR knockout (KO) ferret model and have carried out initial analyses of gross and histologic pathologies in newborn kits ranging from 24-72 hrs of age. The gross presentations in the first two CFTR-KO kits differed significantly. One kit (CF-1) had meconium ileus (MI)-like dilation proximal to an obstruction-like interface, as well as a perforation site more distally in the colon. Distal to the perforation, the intestine appeared to be unusually small, slightly coiled and at the histologic level laden with mucus. The second pup (CF-2) passed stool within the first 24 hrs and had a grossly normal bowel that did not significantly differ in presentation from age-matched WT kits. The variable penetrance of the MI in these CF kits is consistent with the occurrence and severity of MI in CF infants. Gross and subgross examination of CF-1 and CF-2 pancreases did not detect differences from WT as is seen in the CF pig model; however, histological lesions were consistently detected as most acini and ductules were dilated by eosinophilic inspissated zymogen secretions. The histopathology of the CF pancreas suggests that the animals may have at least partial exocrine/pancreatic insufficiency. The lung of CF-1 had no features that were overtly distinguishable from those of WT kits. However, the lung of CF-2 exhibited small foci of bronchopneumonia characterized by degenerative neutrophils that filled bronchioles, and alveoli bearing admixed bacterial colonies. Focal parenchymal necrosis similar to that seen in the case of an infarct was observed, although no thrombus was detected. Historically, bronchopneumonia was not atypical for CF infants (e.g. early 20th century prior to aggressive antibiotic therapies). Bronchopneumonia is also sometimes detected in association with microaspiration of milk in nursing neonatal animals and this may have been the case for CF-2. However, focal bronchopneumonia has yet to be observed in the lungs of control non-CF kits. Histologic analysis of the liver and gallbladder of these CF kits demonstrated these organs were spared from overt disease. Both CF-1 and CF-2 died within 36-42 hrs after birth. The above initial pathology results suggest that CF-1 died of meconium ileus related disease including intestinal obstruction and perforation, and that the cause of death for CF-2 was not readily defined by necropsy but the bronchopneumonia may have contributed. More recently, new cohorts of CF ferrets have been obtained and we are working through pharmacologic approaches to treat both MI and pancreatic disease in the newborns. Thus far, all 12 CF kits have died postnatally and retain diagnostic pancreatic abnormalities. However, approximately 25% (3 of 12 CF kits) appear to escape MI and pass stool. Histologic assessment of the lungs demonstrated recurrent focal bronchopneumonia in a subset of CF kits and we are using aggressive antibiotic therapy to prevent aspiration bronchopneumonia. A mutation in the yeast ABC-transporter, yor1-∆F670 is analogous to human CFTR-∆F508, as evidenced by a similar trafficking defect, retention in the ER, processing of the defective protein by Endoplasmic Reticulum Associated Degradation (ERAD), and failure to reach the plasma membrane. In contrast to the chloride-channel function of CFTR, YOR1 is a drug pump that confers resistance to oligomycin, an inhibitor of mitochondrial function. We constructed a Tet-regulatable allele of yor1-∆F that exhibits titratable resistance to oligomycin, to optimize detection of gene-gene interactions. A novel high throughput method for constructing double mutants was used to introduce Tet-yor1-∆F into the set of 4850 yeast gene deletion strains to screen for enhancers and suppressors of oligomycin resistance and investigate their potential role in trafficking of yor1-∆F to the plasma membrane. For sensitive detection of gene-gene interactions, we analyzed kinetic growth curves for all 4850 double mutant strains at multiple concentrations of oligomycin. Two rounds of screening identified over 100 potential gene-gene interactions. These included several genes having both human orthologs and previously described cellular functions of presumed relevance to CFTR-∆F biogenesis. For example, CDH1, identified as a potential regulator of yor1-∆F in our screen, has been shown previously to function in ubiquitin-mediated protein turnover and the RAS pathway. Another strongly positive gene interaction involved JJJ1, which regulates endocytosis (membrane trafficking) at the cell surface. Neither of these gene products has been implicated previously as part of ABC-protein biogenesis. Followup experiments are in progress to investigate the molecular basis for the phenotypic findings. Once genes involved in post-transcriptional mechanisms are confirmed, the corresponding mutants will be assessed for effects on yor1-∆F protein stability, degradation, and ultimately trafficking to different cellular compartments. The longer-term goal, related to identifying genetic and cellular targets that modulate the biogenesis of yor1-∆F, is to validate evolutionarily conserved roles for functions relevant to biogenesis of CFTR-∆F. Toward this end, we are investigating conserved modifiers using RNAi approaches in established human cell types. In summary, we have developed a yeast genetic model for cystic fibrosis, wherein we hope to scan a eukaryotic genome for extragenic mutations that result from deletion of a conserved phenylalanine residue in NBD1, corresponding to the most common mutation associated with cystic fibrosis. The approach is aimed at identifying new participants in ABC-protein biogenesis, and opportunities for development of novel genetic targets for cystic fibrosis therapeutics. Homologous recombination using recombinant adeno-associated virus (rAAV) has been used to generate new cystic fibrosis models in the ferret and pig. Preliminary evidence from our laboratory suggests that the ∆F508 mutation in ferret CFTR gives rise to a mutant protein that undergoes similar processing defects as seen in human CFTR. Hence, the ferret may be a good species for modeling this mutation. Here we report our current efforts to produced a ∆F508-CFTR ferret model using rAAV mediated-gene targeting in primary fibroblasts and somatic cell nuclear transfer (SCNT) cloning. Our initial strategy involved construction of a rAAV targeting vector (AV-∆F508/neo252) that encoded homologous sequences flanking exon-10 with the ∆F508 deletion and a floxed neomycin selectable marker 252 bp downstream of exon-10. The distance of the Neo selectable marker from exon-10 was chosen to minimize interference with splicing of the CFTR gene. AV-∆F508/neo252 virus was used to infect primary ferret fibroblasts and cell clones were isolated by limiting dilution in 96 well plates under G418 selection and then screened by PCR. Although 0.34% of the screened clones demonstrated homologous recombination at the CFTR locus, none of the clones retained the ∆F508 mutation in exon-10. These findings suggested that the distance between the F508 deletion and the Neo insertion was to great, leading to homologous recombination between the ∆F508 target site and the Neo insertion. We next redesigned the targeting vector (AV-∆508F/neo65) to place the Neo insertion 65bp downstream of exon-10. Of 11,868 clones screened thus far using this vector, 35 clones were positive for a homologous recombination event (0.29%). However, only 15 of the 35 clones demonstrated correct left arm and right arm flanking sequences by PCR. Sequence analysis of exon-10 demonstrated that 12 of the 15 clones with correct flanking sequences contained the F508 deletion and 3 clones remained F508 unchanged. PCR sequencing data also revealed sequence rearrangements in the target site among a significant portion of the targeted fibroblast clones. This low fidelity of homologous recombination, which was not seen when targeting an insertion to fCFTR exon-10, implies mechanistic differences in the recombination events that control large insertions and small deletions. Although the correctly targeted F508-deleted fibroblast clones rapid senesced upon expansion, we were able to amplify enough cells for SCNT from six fibroblast clones and obtain ten mid-gestation embryos for rederival of fibroblasts and Southern blot confirmation of the targeting event. However, fibroblasts derived from these embryos lacked Neo resistance despite the fact they were PCR positive in the initial screen. This was also observed when the exon-10 knockout ferret was generated and is thought to be due to episomal AAV genomes that allow for persistence of non-targeted cells (G418 resistant) in the original gene-targeted clones. We are currently evaluating an altered vector design to increase the fidelity of homologous recombination and attempting to utilize hybrid ITR vectors for gene targeting which minimize circular episomes that interfere with the selection process of clonal fibroblast lines. Cystic fibrosis (CF) is a genetic disease caused by mutations in the CF transmembrane conductance regulator (CFTR). While our understanding of CF disease pathogenesis has significantly improved, chronic pulmonary bacterial infections and inflammation remain the major causes of morbidity and mortality in CF. Because swine share many anatomical and physiological similarities to humans, we recently developed CFTR-/-piglets using recombinant adeno-associated viruses targeting CFTR in pig fetal fibroblasts and somatic cell nuclear transfer to porcine oocytes. At birth, similar to humans these animals demonstrate meconium ileus, pancreatic involvement, and absent epithelial CFTR chloride current. The most common genetic mutation in CF leads to deletion of phenylalanine at amino acid position 508 (∆F508) in CFTR. In humans this mutation disrupts processing of CFTR leading to an incompletely glycosylated protein that fails to properly traffic to the Golgi and plasma membrane. In vitro studies have suggested that some mouse and pig CFTR ∆F508 are fully glycosylated and reach the apical surface. In order to begin to study the biosynthetic defects in CFTR ∆F508 we have generated CFTR-/∆F508 pigs using similar techniques as for generation of CFTR-/-piglets. At birth, all CFTR-/∆F508 pigs had meconium ileus and pancreatic involvement similar to CFTR-/-piglets and humans with CF. Nasal voltage measurements showed lack of epithelial chloride conductance. In differentiated cultures of tracheal epithelia from CFTR-/∆F508 pigs, CFTR appeared primarily as the immature band B form and there was lack of CFTR chloride current. These findings suggest that CFTR-/∆F508 pigs may be a useful model to further investigate the mechanisms underlying the pathogenesis of CFTR ∆F508 processing in CF disease and to develop new therapeutic tools for CF.

Supported by NIH, CFF, and HHMI. The loss of CFTR channel activity leads to cystic fibrosis (CF). Although CF primarily affects the lungs, many CF patients exhibit gut, liver, pancreas, and vas deferens defects, indicating an important role for CFTR outside the lungs. Understanding the pathophysiology of CF in these organs has been difficult due to the absence of a genetically tractable, easily accessible model system. To address these obstacles, we are using zebrafish to examine the role of the CFTR channel during organogenesis. The zebrafish is an optically clear, genetically tractable organism that is amenable to pharmacological and genetic manipulations, allowing live imaging of experi-mental animals. Importantly, CFTR is highly expressed in a variety of zebrafish tissues. To investigate the role of CFTR during organogenesis we are using a variety of organ-specific fluorescent proteins in transgenic zebrafish lines to track the development of individual organs. Inhibition of the CFTR channel during development by pharmacological or genetic manipulations leads to striking morphogenesis defects in the zebrafish eye and endodermal organs. This indicates that CFTR plays a critical role in organogenesis in addition to organ homeostasis. The transparent and genetically tractable zebrafish model provides a powerful system for studying CFTR's role in organogenesis.

This work is supported in part by a grant from the CFF to M.B. Meconium ileus is a congenital obstruction of the intestine that is detected in15-20% of infants with cystic fibrosis (CF) following birth and may present with complicating lesions (e.g. atresia, perforation, etc.) that require surgical intervention. We recently developed a CFTR-/-pig model in which meconium ileus is ubiquitous at birth. In addition to severe obstruction, the CFTR-/-pig has the presence of variable atresia, stenosis, perforation and microcolon that require surgical intervention. Following birth, CFTR-/-and CFTR+/+ (control) pigs were anesthetized with isoflurane and placed on mechanical ventilation. Prior to abdominal surgery, a tunneled central venous catheter was placed in the left internal jugular vein via a cut down approach and brought out between the scapulae. This provided venous access for post-operative care. Next, a left subcostal incision was made and exploration identified the location and severity of the obstruction. Most inspissated meconium was either in the terminal jejunum to ileum or in the spiral colon. At times, the obstruction was severe with atresia and microileum/colon. If this was the case, a loop ostomy proximal to these obstructions was performed. If no impediments to passage of intestinal material were detected or if the obstruction was located in the spiral colon, then the cecum was opened and the inspissated meconium was flushed out of the ileum and colon. Cecostomy formation followed this maneuver. Within hours of the surgery, piglets were freely walking. The pigs are maintained in pICU (pig intensive care unit) care for 2-4 days with intravenous fluids, systemic intravenous antibiotics (metronidazole, ampicillin, and gentamicin), topical wound care, and regular ostomy flushing. The pigs remain in pICU care until they have successfully transitioned back onto milk-based nutrition and hydration with normal outflow of ostomy material. Surgical correction for meconium ileus obstruction in neonatal pigs is now routine and permits long-term survival of the CFTR-/-pig for study of CF disease pathogenesis. Delivering the cystic fibrosis gene (CFTR) to the lungs of cystic fibrosis (CF) patients has proven to be a significant challenge. To select a lead gene therapeutic for CF lung disease we undertook a systematic candidate selection by screening novel adeno-associated virus (AAV) vectors in defined models of airway epithelium. Our work led to the identification of a lead AAV vector candidate known as AAV6.2 that has a superior transduction profile in mouse and macaque lung in vivo; human and macaque airway in vitro. In efforts to progress AAV6.2 into the clinic, we evaluated its immunogenicity in mouse lung. AAV6.2 expressing β-galactosidase (nLacZ) was intratracheally delivered into C57BL/6 and BALB/c mouse lung, and T cell responses were evaluated at the peak of nLacZ expression. The IFN-γ ELISpot assay confirmed the presence of a high T cell response to both the vector and the nLacZ gene product in lung. Longitudinal gene expression studies demonstrated that although nLacZ-specific T cells were present in lung, nLacZ expression was not impacted, suggesting that the cytolytic activity of these T cells was suppressed. In an attempt to identify the mechanisms underlying the persistence of gene expression in airway we focused on the role of alveolar macrophages (AMs). It is long known that resident AMs are the main reason why lungs remain sterile and clean despite daily encounters with noxious particles in inspired air. Using clodronate liposomes we depleted AMs from BALB/c mouse lung prior to the intratracheal delivery of the AAV6.2 vector expressing nLacZ. We evaluated nLacZ expression in lung and monitored T cell responses to both the vector and nLacZ in lung and spleen up to day 90. We also evaluated B cell responses to vector and nLacZ in serum and the bronchoalveolar lavage fluid (BALF). At day 7, while the magnitude of the vector-and nLacZ-specific T cell response in spleen was similar between the AM depleted and the naïve animals; in lung there was a significantly higher T cell response in the AM depleted mice. Similarly, IgG and IgA responses to the vector and LacZ were significantly elevated in the serum and BALF of the AM depleted mice. Importantly, at days 14, 21 and 90 nLacZ expression in the airways of the AM-depleted mice was ablated and was accompanied by high levels of vector-and nLacZ-specific T cells as well as increased IgG and IgA responses to both vector and nLacZ in serum and BALF, respectively. To summarize, our data show for the first time the role of resident AMs in mediating tolerance to both the AAV vector and transgene product delivered to the lung. Findings in CFTR knockout mice using AAV6.2 vector expressing CFTR are also being discussed.

Supported by the CFF R881, P01-HL051746, CFF LIMBER08G. Although recurrent lung infection and inflammation is the leading cause of morbidity in cystic fibrosis (CF), the role of CFTR in the modulation of the innate immune response of respiratory epithelia remains controversial. While the presence of increased level of neutrophils, free neutrophil elastase activity and IL-8 in the bronchoalveolar lavage of CF infants without the evidence of infection favours the suppressive role of CFTR in pro-inflammatory cytokines secretion, the absence of infection remains under debate. Cell models yielded contradictory results that could be, in part, attributed to the impact of genetic background variability, modifier genes and genetic instability of clonally-selected immortalized cell lines. To assess the role of CFTR expression in the innate immune response of respiratory epithelia in vitro and to minimize the potentially confounding genetic influence, we expressed the human CFTR under the control of an inducible transcriptional activator in the CF bronchial epithelia CFBE41o-. This experimental system allows the gradual adjustment of the channel expression to a level comparable with the endogenous expression in Calu-3. To mimic the air-liquid interface environment of lung epithelia, the cells were cultivated on semipermeable filter support under either air-liquid (ALC) or liquid-liquid (LLC) culture conditions. Under LLC condition we did not observe differential cytokine secretion regardless of the CFTR expression level. In contrast, under ALC condition CFTR expression significantly suppressed the increased secretion of pro-inflammatory cytokines, including IL-6 and IL-8, observed both in the apical and the basolateral compartment. Parental CFBE41o-cells, similar to a cell line harbouring inducible expression of the folding-deficient ∆F508 CFTR, also secreted significantly higher levels of IL-6 and IL-8 compared to wt CFTR expressing epithelia. Initial studies using quantitative RT-PCR analysis on a panel of unfolded protein response markers, including X-box protein 1 (XBP1), C/EBP homologous protein (CHOP), or Erdj4 and Western blotting using antibodies raised against UDP-glucose:glycoprotein glucosyltransferase (UGGT) or the phosphoeukaryotic initiation factor 2α (eIF-2α), suggested that the increased cytokine secretion is independent of the generation of an ER-stress response. Even upon induction of ∆F508 CFTR expression there was only negligible increase in the ER-stress response, suggesting the involvement of other signalling mechanism in the pro-inflammatory cytokine secretion in CF respiratory epithelia. Despite the importance of nasal potential difference (NPD) as an in vivo biomarker of CFTR activity, a single, unified protocol has not been adopted worldwide. Moreover, performance of the assay in multi-center clinical trials has been associated with unreliability, excessive artifact, and inter-center variability that may adversely affect sensitivity. Following an analysis of NPD methods, we successfully pilot-tested a "Universal" NPD protocol utilizing electronic data capture (EDC-with human grade bioamplifier) and use of a non-perfusing (agar) nasal catheter less susceptible to artifacts. To confirm the validity and feasibility of the agar approach, we performed a multicenter, randomized, crossover trial to compare previous methodology (site specific standards in accordance with TDN SOP using a continuously perfusion catheter) and a proposed agar-based NPD protocol. Normal subjects randomly underwent two-nostril NPD with each method, separated by ≥ 1 day. We enrolled 22 subjects in 6 centers, and all completed paired analysis. PD values measuring total chloride conductance (TCC) were significantly more polarizing using the agar method (-24.2 ± 12.9 mV agar vs. -18.2 ± 9.1 mV with perfusion; p<0.05). Similar trends were seen in individual measures of chloride conductance (change zero chloride: -12.7 ± 10.4 mV agar vs. -8.7 ± 5.6 mV perfusion, p<0.05; change isoproterenol: -11.5 ± 6.2 mV agar vs. -9.5 ± 6.3 mV perfusion, p=0.15). Results may reflect a more potent Clgradient due to the absence of continuous Ringers with the agar method. Despite an enhanced signal, no increase in CV was observed (0.53 agar vs. 0.50 perfusion for TCC). Agar tracings were significantly more reliable, demonstrating 3.5 and 1.5 artifacts per tracing in the sodium and chloride components, respectively, compared to 11.4 and 8.8 with the perfusion method. Tracing stability (estimated by the variability of the measure over a 10 sec interval) was significantly enhanced with the agar method (0.42 vs. 0.71 with perfusion). Limited comparison with an alternative non-perfusion device (ECG with AgCl electrodes) showed similar beneficial results. Operators at all sites preferred the agar method, and reported reduced setup and troubleshooting. Interim results shared with CFFT and industry partners endorsed use of the approach, which will be used in two upcoming international trials examining CFTR modulators. In conclusion, the proposed non-perfusion NPD approach is preferred over traditional (perfusion) NPD methods, and may improve sensitivity to changes in Clconductance within therapeutic trials. The proposed NPD procedure should be considered for adoption for therapeutic monitoring worldwide.

Supported by CFFT and NIH. Background: Airways in cystic fibrosis (CF) are characterised by a reduction in water content of the airway surface liquid (ASL) and accumulation of thick mucus, leading to chronic infection, inflammation and tissue damage. Mannitol is an osmotic agent that increases water content of the ASL, improves mucociliary clearance and in short term studies has been shown to improve FEV 1 (1, 2, 3) . This longer-term phase III study was designed to assess efficacy and safety of inhaled mannitol.

Methods: The study was a randomized, double-blind, placebo-controlled design. Subjects inhaled 400 mg mannitol or placebo bd for 26 weeks and then had the option to receive active treatment in an open label phase. Patients had confirmed CF, were > 6yrs with an FEV 1 of 30-90% predicted.

Study Objectives: The primary endpoint was change in FEV 1 after 26 weeks. The study was also powered to assess change in FEV 1 in patients taking rhDNase. Additional endpoints included pulmonary exacerbations and antibiotic treatment rates.

Results: Baseline characteristics for the 295 patients were as follows: mean age 23 ± 11.3yrs, 44.7% female, % predicted FEV 1 62.0 ± 16.3%, and 55.3% of patients were concurrently treated with rhDNase.

Patients treated with mannitol had a clinically significant improvement in FEV 1 from baseline of 122 mL after 26 weeks (p<0.001 vs. placebo). This improvement in lung function was shown at week 6 and was sustained through to week 26 ( Figure 1 ). For patients on mannitol and concurrent rhDNase, FEV 1 improved after 26 weeks by 96.2 mL (5.2%) from baseline (p=0.002 vs. placebo). There was significant improvement in FEV 1 in patients treated with both mannitol and rhDNase (p=0.008 vs. placebo), and in those being treated with mannitol alone (p=0.015 vs.placebo).

Mannitol was well-tolerated with similar rates of adverse and serious adverse events in the groups. The most common adverse event was cough, mild to moderate in most cases, and similar between the treatment arms.

In patients with CF, mannitol improves lung function and appears to have an acceptable safety profile. Cystic fibrosis (CF) is a genetic disorder caused by defective CFTR function leading to abnormal epithelial fluid transport in the lungs and other organs. VX-770 is a CFTR potentiator under clinical investigation in CF. This randomized, double-blind, placebo-controlled study was designed to evaluate VX-770 safety and tolerability; clinical outcomes (spirometry and CFQ-R) were also assessed. Subjects were adults with CF and a G551D-CFTR allele. Part 1 was a cross-over design in which 20 subjects received placebo or VX-770 q12h at 25, 75, or 150 mg for 14 days. Part 2 was a parallel design in which 19 new subjects received placebo or VX-770 q12h at 150 or 250 mg for 28 days. In both parts, no drug-related serious adverse events were reported and VX-770 was generally well-tolerated. Results demonstrating statistically significant improvement in forced expiratory volume (FEV1) from baseline to days 14 and 28 have previously been reported. Changes in forced vital capacity (FVC) and forced expiratory flow (FEF25-75) have now also been analyzed. In Part 1, mean (±95%CI) In the analysis of respiratory outcomes by CFQ-R in Part 2, median improvement in respiratory domain score at Day 28 in the 150 and 250 mg groups was 8.3 (p=0.06) and 11.1 (p=0.08), respectively. In the 250 mg group, 6 of 7 subjects at 14 days, and 5 of 7 at 28 days demonstrated a minimal clinically important difference in their respiratory domain score (≥4 points). Further details of effect of VX-770 on CFQ-R by domain will be presented. Conclusion: 14-and 28-day administration of VX-770 led to improvement in early respiratory outcomes and further studies are needed to assess persistence of effect. Certain aminoglycosides (AG) can induce "translational readthrough" of premature termination codons (PTCs) in CFTR, and the approach has been shown to restore full-length functional protein in a number of preclinical and clinical settings. However, cumulative toxicity and relative lack of efficacy of AGs deter development as a treatment for genetic disease caused by PTCs, including the ~12% of affected cystic fibrosis (CF) patients. Using a structure-based approach, a series of synthetic AGs (termed NB30 and NB54) were developed that exhibit enhanced suppression of PTCs using a dual-luciferase readthrough construct while also exhibiting reduced toxicity in cellular assays. To confirm the utility of the approach in CF, we tested NB30 and NB54 in comparison to gentamicin (gent). In a fluorescencebased halide efflux assay (SPQ), all agents (500 mcg/mL incubated 18 hrs) restored activity of W1282X CFTR expressed in HeLa cells (p<0.01), with a relative rank-order of NB54, NB30, and gent. Confirmatory testing in an SPQ-based plate-reader assay of IB-3 cells (endogenous genotype W1282X/∆F508) indicated dose-dependent rescue of CFTR activity by NB30 and NB54 at the highest concentrations tested (2.4 and 2.2 fold over control at 1000 mcg/mL, respectively; p<0.001), whereas gentamicin-treated cells peaked at the lowest tested dose (125 mcg/mL). As expected, no rescue of CFTR-mediated halide efflux was seen with CFNPE cells (endogenous ∆F508/∆F508) or HeLa parental (no CFTR transgene) cells. To confirm the effect in polarized epithelia, we assessed forskolin-stimulated Isc in CFBE41o-cells stably transfected with W1282X. Treatment of monolayers with NB54 for 18 hours resulted in 0.526 µA/cm 2 of CFTR dependent Isc over vehicle, whereas gentamicin induced 0.084 µA/cm 2 (p<0.05 vs. gentamicin, n=12). Dose-dependent comparisons in CFBE41o-cells transiently infected with G542X CFTR showed similar results. Limited studies utilizing primary human nasal airway epithelial cells expressing 2789+G->C/DF508 (heterozygous for a splice mutant encoding an in-frame PTC) treated with NB54 (500 mcg/mL) indicated activation of forskolin-mediated CFTR transport (Isc=2.4 µA/cm 2 vs 1.5 µA/cm 2 with vehicle). Preliminary results in vivo indicate that systemic administration of NB54 (120 mg/kg SC) compared to gentamicin (30 mg/kg, SC) to CF -/-hCFTR-G542X transgenic mice for 2 weeks restored comparable chloride transport following forskolin addition in excised murine intestines studied under Isc conditions (23 µA/cm 2 vs. 24 µA/cm 2 , respectively), but greater chloride transport with NB54 following sequential carbachol addition (47 µA/cm 2 vs. 22 µA/cm 2 , respectively), suggesting more robust CFTR rescue. When taken together, these results indicate that the synthetic aminoglycoside NB54 demonstrates enhanced suppression of PTCs within CFTR compared to gentamicin. Synthetic aminoglycosides may provide an improved means of treating genetic diseases caused by PTCs, and deserve further exploration as therapeutic agents in nonsense mediated CF.

Supported by the CFF and the NIH. FoldRx Pharmaceuticals has developed a powerful yeast-based, highthroughput platform designed to identify small-molecule modulators of protein trafficking. Many of the proteins and cellular mechanisms involved in protein trafficking and quality control in yeast have similar counterparts in humans. Proof-of-concept studies demonstrated that modulating endoplasmic reticulum (ER) to Golgi transport in yeast could be used to identify small molecules that correct the ∆F508 defect in mammalian cells. Several known correctors demonstrated significant activity in the yeast trafficking assay.

We previously reported on the activities of a lead series discovered through a limited and directed yeast trafficking screening assay that have shown dual corrector and potentiator activity in primary epithelial cultures from CF patients. This chemical series is under active development as a therapy for cystic fibrosis. We detail here the results of de novo screening of compound libraries using a trafficking assay in yeast to discover additional correctors of ∆F508 CFTR.

A diverse library of 250,000 compounds was assayed for effects on trafficking in yeast. A collection of 5,200 compounds including hits from the screen and a number of structurally diverse inactive analogs was tested for ability to correct the ∆F508 defect using ion flux assays in FRT cells expressing recombinant ∆F508 CFTR and halide-sensitive yellow fluores-cent protein. Nine families of chemically tractable compounds were identified. Several families of compounds possessed dual activity, acting as both correctors and potentiators. These findings demonstrate the usefulness of trafficking assays in yeast to discover novel ∆F508 rescue compounds.

Assays for corrector activity in FRT and A549 cell lines have been shown to give different rankings of corrector activity in an evaluation of a small set of known correctors. Compounds active in FRT cells were profiled in assays for correction in A549 cells and a subset of compounds showed activity in both cell lines. These compounds are being evaluated in CF primary epithelia (CF hBE) and initial experiments show promising activity.

The mechanism of action of these compounds is being actively investigated in both yeast and mammalian cells. Unbiased approaches based on characterization of genes that suppress or enhance the yeast phenotype as well as compound-and hypothesis-driven strategies are being employed. These novel compounds have the potential to benefit cystic fibrosis patients by reversing the basic defect associated with the ∆F508 mutation.

This work was supported by the Cystic Fibrosis Foundation Therapeutics, Inc. through a research, development, and commercialization agreement. Experiments in CF hBE were performed at ChanTest, Inc. Activation of alternative chloride channels by drugs is one strategy for cystic fibrosis (CF) therapy, with two compounds (denufosol and Moli1901) in clinical trials that activate calcium-activated chloride channels (CaCCs) by transient elevation of cytoplasmic calcium. We reasoned that compounds that directly target CaCCs might be superior to non-selective calcium-elevating agonists in terms of fewer off-target effects and the possibility of producing sustained rather than transient CaCC elevation. TMEM16A was found recently to be a CaCC expressed in airway epithelial cells. A highthroughput screening (HTS) assay was developed based on iodide quenching of YFP fluorescence in FRT cells stably expressing human TMEM16A. The HTS assay was designed to identify TMEM16A inhibitors and activators in the same kinetic measurement. Screening of 100,000 diverse, druglike small molecules yielded many active TMEM16A inhibitors, with several classes having nanomolar potency and inhibiting CaCC chloride conductance in multiple cell types of airway, salivary gland, pancreas and intestinal origin. A smaller number of TMEM16A activators were identified, with at least 5 classes strongly activating TMEM16A with submicromolar potency. Activation was independent of calcium chelation and without cytoplasmic calcium elevation, suggesting that the compounds target TMEM16A directly. The compounds produced a sustained increase in chloride current in short-circuit current measurements on primary airway epithelial cells from CF subjects. Small-molecule activators of TMEM16A and related CaCCs may be useful in the therapy of CF.

Lymp, J. 1,2 ; Hilliard, K. 3 Pediatr Pulmonol Suppl 2005; 28:125) , one of the two co-primary endpoints was pulmonary exacerbation, which was determined via a respiratory and systemic symptoms questionnaire (RSSQ) specifically developed for the trial by Boehringer Ingelheim.

Methods: The RSSQ was developed with the intent of capturing the signs and symptoms that constitute the definitions of a pulmonary exacerbation of Fuchs et al. (NEJM 1994; 331: 637-42) and Rosenfeld et al. (J Pediatr 2001; 139: 359-65) . The instrument allows for the collection of information regarding 14 signs and symptoms from both definitions, and was implemented in standardized interview format to reduce variability. Moreover, the RSSQ was translated into 9 languages for international use. The instrument was administered at a pre-treatment screening visit, every 4 weeks for the 24 week treatment duration, and also 4 weeks post-treatment for a total of 9 measurements. These signs and symptoms were combined with physical findings, lung function and intravenous antibiotic use to determine the presence of pulmonary exacerbation using the definitions of Fuchs et al. and Rosenfeld et al.

Results: The trial was stopped early due to a higher rate of serious adverse events in adults on active treatment. At the time of study termination, there were 420 subjects randomized, with 164 (39.0%) of these completing the 24 week questionnaire; in total, 1946 questionnaires were completed during the treatment period. The most common symptom was increased frequency of cough, reported by 325 (77.3%) subjects. Other symptoms reported by at least 60% of the subjects were increased sputum production (71.0%), increased intensity of cough (67.6%), increased malaise, fatigue or lethargy (62.1%), and increased chest congestion (62.1%). During the treatment period, 271 (64.5%) subjects met the Fuchs signs and symptoms component (not incorporating IV antibiotic use), while 337 (80.2%) subjects met the Rosenfeld signs and symptoms definition (not incorporating FEV 1 ), 109 (26.0%) subjects experienced at least one hospitalization and 130 (31.0%) subjects experienced at least one course of IV antibiotics.

Conclusions: The RSSQ can be used to capture both the Fuchs and Rosenfeld definitions of pulmonary exacerbation, and also to develop new definitions and evaluate each sign and symptom individually. In this trial, changes in signs and symptoms occurred in more subjects than other pulmonary exacerbation related outcomes, indicating that use of the RSSQ might allow for a more sensitive definition of pulmonary exacerbation.

Acknowledgments: Supported by Boehringer Ingelheim and Cystic Fibrosis Foundation Therapeutics, Inc. Nasal potential difference (NPD) is a measure of CFTR and ENaC activity used in early phase clinical trials, but it remains unclear which NPD parameters are most sensitive to changes in CFTR activity. Relating NPD measurements and clinical outcomes (eg, lung function, patient reported outcomes) can help optimize the design of trials that utilize NPD and aid in prioritizing drug candidates for testing in larger trials. In a two-part safety and tolerability study in patients with the G551D CFTR mutation, the oral CFTR potentiator VX-770 was generally well-tolerated with no drug-related serious adverse events. This study also included biomarker and clinical efficacy endpoints. NPD was performed throughout the study, affording an opportunity to explore relationships between clinically meaningful outcomes and nasal ion transport and providing important reproducibility data. A NPD parameter measuring CFTR-dependent Cltransport (change in PD following zero Clsolution and isoproterenol perfusion) demonstrated a significant change from baseline within the treatment group for VX-770 for the 75 and 150 mg doses at Day 14 in Part 1 (-4.7 and -5.4 mV, respectively; P<0.05) and VX-770 150 and 250 mg doses at Day 28 in Part 2 (-3.5 and -5.5 mV, respectively; P<0.05). The corresponding changes in the placebo groups at those time points were -1.7 and -0.4 mV, respectively. Improvement in Cltransport was paralleled by improvements in NPD measures of Na + ion transport, such as change from baseline to Day 14 (Part 1) in amiloride response (-3.5 and -8.6 mV for the VX-770 75 and 150 mg groups vs +4.5 mV in the placebo group; P<0.05 vs placebo) though Na + transport changes from baseline were not significant at Day 28 (Part 2). Improvements were also observed in sweat chloride biomarker and lung function endpoints in this study. Correlations between zero Clplus isoproterenol to sweat chloride were weak for all treatment groups (Pearson coefficients for VX-770 150mg: -0.15 after 14 days and -0.23 after 28 days in combined analysis of Parts 1 and 2 of the study). Correlations of zero Clplus isoproterenol to lung function (%predicted FEV1) after 14-and 28-days of therapy were similarly weak (+0.27 after 14 days and -0.32 after 28 days). Inconclusive correlations were probably due to the small sample size. The results of this small study suggest that oral administration of a CFTR modulator compound can improve NPD measures of CFTR-related Cltransport and may lead to an improvement in ENaC-related Na + transport. In addition, though NPD-measured ion transport, sweat chloride, and lung function all showed improvement in this study, the changes were not necessarily correlated. Further clinical evaluation will be necessary to better understand the relationship between biomarkers and outcome measures. Introduction: Sensitive outcome measures to assess the efficacy of therapeutic interventions in cystic fibrosis (CF) patients with mild lung disease are currently lacking. We studied the ability of the Lung Clearance Index (LCI), a measure of ventilation inhomogeneity determined during Multiple Breath Washout (MBW) to detect a treatment response to hypertonic saline inhalation in pediatric CF patients with normal pulmonary function.

Methods: In a double-blind crossover trial twenty CF patients received 4 weeks of hypertonic saline (HS) and isotonic saline (IS) in a randomized sequence separated by a 4 week washout period. The primary endpoint was the change in LCI in HS versus IS treated patients. This trial was registered with ClinicalTrials.gov, number NCT00635141.

Results: Four weeks of twice daily inhalation of HS significantly improved the LCI as compared to IS (mean difference 1.04 (SD 0.94), p=0.0004). Baseline LCI before IS, 8.91+/-2.40 was not significantly different from baseline LCI before HS inhalation, 8.79+/-1.91 (p=0.64). Randomization order had no significant impact on the treatment effect. No significant changes were observed in secondary outcome measures. (FEV 1 % predicted, FVC% predicted, FEF 25-75 % predicted, CFQ-R respiratory domain score and CFQ-R parent respiratory domain score.) LCI negatively correlated with all secondary outcome measures: FEV 1 % predicted (r=-0.61, p=0.0001), FVC% predicted (r= -0.45, p=0.0001), FEF 25-75 % predicted (r=-0.53, p=0.0001), CFQ-R respiratory domain score (r= -0.43, p=0.0001) and CFQ-R Parent respiratory domain score (r= -0.34, p=0.0082). The frequency of adverse drug reactions was similar for both HS and IS inhalations (5 versus 7, p=0.17) .

The LCI was able to detect a treatment effect from HS inhalation in CF patients with mild disease and may be a suitable tool to assess early intervention strategies in this patient population. ON, Canada; 4. Diagnostic Imaging, North York General Hospital, Toronto, ON, Canada Rationale: Neutrophils play a key role in the pathogenesis of cystic fibrosis (CF) lung disease. However our current modalities to measure airways inflammation are limited. 18 FDG PET-CT is a promising tool because it is non-invasive and has been shown to quantify and localize neutrophilic inflammation (1, 2) . Our goal is to show that 18 FDG PET-CT is able to detect changes in inflammation that result from intravenous antibiotics for a pulmonary exacerbation in pediatric CF patients.

Methods: Pediatric CF patients ranging between the ages of 6 and 18 years admitted to Sick Kids Hospital for a pulmonary exacerbation were eligible for the study. 18 FDG PET-CT scans (with low dose CT) were performed on days 1 and 14 (+/-72 hours) of admission. 18 FDG PET-CT scans from 10 age-matched oncology patients without active pulmonary disease acted as controls. Standard uptake value (SUV) was the form of analysis. The lungs were divided into four zones for analysis: right upper, right lower, left upper and left lower. Background activity (SUVmean) and superimposed focal uptake (SUVmax) were measured in each lung zone. The SUV values from the 4 zones were averaged to generate the SUVmean and summed for the SUVmax.

Results: Ten CF patients have been enrolled in this study to date. The mean (SD), range for age was 11.3 (3.9), 6-16 years. Six of the ten patients were female, five were Pseudomonas aeruginosa positive, and all were pancreatic insufficient. The mean (+/-SD) of FEV 1 % predicted day 1 (pre-therapy) and day 14 (post-therapy) were 59.6% (+/-14.0) and 75.6% (+/-13.5), respectively (p=0.0001, paired t test). Both pre-therapy SUVmean and SUVmax were significantly different from controls (p values ). SUVmean in CF patients significantly improved with intravenous antibiotic therapy (mean difference +/-SD) of 0.19 +/-0.099, p<0.002), but post-therapy SUVmean was not different (p=0.38) from controls. Antibiotic therapy resulted in a significant decrease in SUVmax (mean difference +/-SD: 1.83 +/-0.81, p<0.001) and the difference between post-therapy SUVmax and controls trended towards significance (p=0.053) with focal uptake being absent in controls. Improvements in FEV 1 in percent predicted from baseline were negatively correlated to changes in SUVmax during therapy. (R=-0.59, p=0.007).

Conclusions: 18 FDG PET-CT is able to detect changes in inflammation that result from treatment of a pulmonary exacerbation in pediatric CF patients. Analyzing focal inflammation (SUVmax) may be a more sensitive measure of airways inflammation than overall 18 FDG uptake in CF patients. 1995;332:848-54) . However, IBU use is suboptimal due to rare, but dramatic side effects (Konstan et al. Curr Opin Pulm Med 2008; 14:567-73 ). There-fore, an alternative anti-inflammatory agent with a better safety profile is needed. Using a surrogate biomarker for neutrophil (PMN) migration to the airway mucosa, delivery of PMNs to the oral gingival mucosa (Wright et al. Blood 1986; 67:1023-30) , we previously showed that high-dose IBU limits the delivery of PMNs (Konstan et al. J Pharmacol Exp Ther 2003; 306:1086-91) . We chose 2 other drugs with known anti-inflammatory properties, simvastatin (SIM) and pioglitazone (PIO), to evaluate as potential alternatives to high-dose IBU using this mouthwash model.

Methods: Healthy subjects (n=25; 19-46 yrs; 14 females) free of gingival disease were randomized 2:2:1 to oral PIO (30 mg once/day, n=10), SIM (40 mg once/day, n=10), or IBU (1000-1600 mg twice/day, to achieve Cmax 50-100 µg/mL; n=5) respectively for 10 days. PMN counts were determined during 3 time periods: Baseline (B: days 1-3), Treatment (T: days 3-10) and Recovery (R: days 13-15), on days 1, 2, 3 (B); 8, 9, 10 (T); and 13, 14, 15 (R). PMN counts were measured from mouthwashes: following an overnight fast, each healthy subject swished normal saline (20 mL) for 30 seconds then returned the specimen to a cup, with a second mouthwash performed after the first. Specimens were centrifuged (1000xg, 4∞C) for 10 min. Cell pellets were re-suspended in Hank's buffer containing 3 µg/mL of acridine orange (fluorochrome). Re-suspended cells were incubated for 15 min (37∞C), and PMNs easily counted by fluorescence microscopy. Data from the second daily mouthwash was analyzed as it best represents more immediate PMN recruitment and has less day-to-day variabilty.

Results: Change from baseline to end of treatment in mean oral PMN counts for the PIO, SIM, and IBU groups was +6.4%, -19.6% and -28.4% respectively. The change for the IBU group was significant in a paired t-test (p=0.048), where four of five subjects had decreased PMN counts. This is also consistent with the 25.0% decrease reported for IBU in healthy subjects from our previous study (Konstan et al. J Pharmacol Exp Ther 2003; 306:1086-91) . Six of 10 subjects in the SIM group had PMN decreases, while only four of 10 subjects in the PIO group decreased.

There were 14 adverse events during the study, which were judged insignificant. Joint pain in 2 subjects were judged possibly related to SIM.

Conclusion: Similar results for IBU in this and our previous study demonstrate reproducibility of this mouthwash test as a non-invasive outcome measure for PMN delivery to a mucosal surface. We speculate that the dose of PIO may have been too low to have an effect on PMN in this model. Further study of both SIM and PIO seems warranted.

Supported by the Cystic Fibrosis Foundation. Markers obtained by rapid and non-invasive methods could be used in the precocious detection of exacerbations or as prognostic indicators in cystic fibrosis (CF). Our goal was to search for lipid signatures characteristic of CF patients in blood plasma using two separate mass spectrometry approaches: a novel MALDI-TOF-ClinProTools analysis, and multiple reaction monitoring (MRM) LC-ESI-MS/MS. Samples from 53 CF patients and 18 healthy children were collected at the Institute for Mother and Child (Warsaw). Organic extraction was followed by column chromatography separation of lipid classes. Extracts were analyzed by mass spectrometry (MALDI-TOF), ion signatures were compared by the ClinProTools software (Bruker Daltonics) and relevant peaks were identified by LC-ESI-MS/MS. The most remarkable differences were found in the phospholipid class. Eighteen ions were significantly decreased in CF patients, corresponding to 6 lysophosphatidylcholine (LPC16:0, 18:0, 18: 1, 18:2, 20:3, 20:4) and 12 phosphatidylcholine species (PC32: 1, 34:2, 36:1, 36:2, 36:5, 38:0, p38:3, 38:4, 38:5, 38 :6, e40:0, p40:1). One ion (either LPCe16:0 or LPE18:0) was significantly increased in patients. Five ions (LPC20:3 and PC36: 1, 38:3, 38:4, 40:6) were consistently downregulated in severe vs mild patients (according to Shwachman score). In parallel, organic extracts from 16 healthy children, 29 mild and 6 severe patients were subjected to LC-ESI-MS/MS analysis and quantification. This approach found 3 peaks sig-nificantly downregulated in patients vs controls (LPC18:0, PC32:1 and PC40:6), confirmed 4 of the 5 differential peaks found by the first approach between mild and severe, as well identified 3 PC (36:0 or p38:6, 36:2 and 36:3) and one sphingomyelin form (SM24:1) decreased in severe patients. These results indicate that some plasma phospholipid signatures may be able to discriminate mild and severe forms of CF, and shows for the first time that MALDI-TOF and ClinProTools may be a suitable methodology for the search of lipid markers in CF. EPIX is developing novel small molecules that can correct the folding defect of the ∆F508 CFTR mutant, leading to increased cell surface expression and accordingly enhance CFTR function. We have previously presented an EPIX-generated full length model of wild type CFTR, from which a model of the intracellular domains of ∆F508 CFTR had been generated. With this model a total of five in silico screening campaigns were carried out that targeted four different sites in the domain-domain interfaces of ∆F508 CFTR. In parallel, a ligand-based pharmacophore screen was conducted. A prioritized set of 100 to 300 compounds from each in silico screen was purchased and tested in biological assays. Active hits were identified in a highthroughput assay employing FRT cells that stably express the human ∆F508 CFTR and the halide-sensitive yellow fluorescent protein (Galietta et al. Am J Physiol Cell Physiol 2001; 281:C1734-42) . Hits were confirmed in Ussing chamber assays with FRT cells expressing ∆F508 CFTR. Short circuit current changes were recorded in response to subsequent additions of 10 µM forskolin, 100 µM IBMX and 20 µM genistein.

A compound identified from the pharmacophore screen was the starting point for the most advanced lead series. As previously described, this series is unique from published CFTR modulators because the compounds have dual corrector-potentiator activity. Based on our FRT cell Ussing chamber results, we have improved the corrector efficacy ~6-fold (24 h incubation with 10 µM compound) from the first lead compound, and these compounds achieve ~3-fold more correction over basal activity than corrector 4a (6 µM). The additional potentiator activity helps to boost the forskolin-induced current to 10-fold of that observed for vehicle-treated control cells. Compound activity has been confirmed in Ussing chamber experiments with primary CF human bronchial epithelial (hBE) cells. In these cells the best compounds to-date achieve forskolin-induced currents that are ~3-fold larger than vehicle-treated cells. Equally, the current that can be inhibited with 10 µM CFTR(inh)-172 is 2-3 fold greater than that from vehicle-treated cells. Increased Band C formation was observed in cell surface biotinylation experiments after 24 h incubation with compound using HeLa cells expressing ∆F508 CFTR; thus confirming the compound-mediated correction of the folding defect of ∆F508 CFTR. The substantial improvements in compound activity have achieved efficacy levels close to what is considered necessary for beneficial effects in ∆F508 CF patients.

This work was conducted in collaboration with ChanTest, Inc. (Cleveland, OH, USA) and is supported and funded by CFFT. Thomas, D.Y. 1 1. Biochemistry, McGill University, Montreal, QC, Canada; 2. Chemistry and Earth Sciences, University of British Columbia, Vancouver, BC, Canada; 3. Physiology, McGill University, Montreal, QC, Canada; 4. Biochemistry, Erasmus University, Rotterdam, Netherlands; 5. Cell Biology, Erasmus University, Rotterdam, Netherlands Cystic fibrosis (CF) results from mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene. The most frequent mutation, deletion of phenylalanine 508 (DF508), causes a folding defect that inhibits trafficking to the plasma membrane. DF508 retains some anion channel function if it reaches the cell surface, which makes it an attractive target for pharmacotherapeutics. We constructed a novel high-throughput (HTS) cell based assay to test the ability of small molecules to restore DF508 CFTR trafficking to the plasma membrane. As marine sponges are a rich source of natural products having novel chemical structures and potent biological activities, we investigated 720 sponge extracts for their ability to correct DF508 trafficking. After 6 rounds of purification and HTS assay, we isolated three closely related members of the latonduine family of alkaloids having corrector activity from the Stylissa group of sponges. These compounds increased CFTR surface expression up to 80% with detectable channel activity at 1 nM concentration. Results were confirmed in the airway epithelial cell line CFBE, which is derived from a cystic fibrosis patient homozygous for DF508. Latonduine A was tested further in vivo using a mouse salivary secretion assay. CFTR-dependent secretion increased 4-fold when CF mice received 0.25 mg/kg latonduine A twice daily for 2 days by intraperitoneal injection. A biotinylated analog of latonduine A was prepared to identify molecular targets by affinity purification and mass spectrometry. Reducing the expression of several putative targets by siRNA knockdown led to partial rescue of functional CFTR. These compounds and their potential target proteins are currently under active investigation.

Supported through the BREATHE program funded by the CCFF and CIHR. Small molecule agents that overcome the primary ∆F508-CFTR processing defect (CFTR Correctors) are being pursued as therapies for cystic fibrosis (CF). Several such compounds from high-throughput drug screening (HTS) programs are available for study through the CFFT Modulator Library (R.J. Bridges, Rosalind Franklin University). We previously reported a comparison of the first four corrector agents available through the library, indicating a rank order of Corr-4a>VRT-325=VRT-640≥Corr-3a by short-circuit (Isc) assays of CFTR Clchannel function. We have since extended this study to 13 additional compounds, using forskolin (20 µM)and genistein (50 µM)-stimulated Isc to evaluate ∆F508-CFTR activity at the plasma membrane. Experiments were conducted in FRT and CFBE41omonolayers stably transduced with ∆F508-CFTR. In ∆F508 FRT monolayers, the cell type frequently used in large-scale HTS programs, ∆F508 correction rank order was observed as follows: (Corr-4c, -15Jf, -4d = Corr-4a)>(Corr-2b, -2i, -5a, -5c, dynasore = low temperature (27 o C x 48 hrs))>(Corr-3d, cmpd-48, KM11057 = vehicle control)>(C8 and KM11060) (p<0.05 between groups). In CFBE41o-cells using an identical Ussing chamber protocol, rank order was similar except for low temperature (as previously reported): 27 o C >(Corr-4c, -4d)>(all other compounds = Corr-4a = vehicle control)>(C8 and KM11060) (p<0.05). Three of the most active correctors, Corr-4a, -4c and -4d, are all bismethylbithiazoles. Interestingly, a fourth compound of that class, Corr-15Jf, was active in the more permissive FRT model but not CFBE41o-, reinforcing that important differences exist between commonly utilized cell models. In FRT cells, each compound augmented both forskolin-and genistein-induced Isc, with the exception of Corr-4c and -4d, which only increased forskolin response (p<0.05 vs. vehicle control). In CFBE41o-(known to be relatively refractory to cAMP-based activation) treated with Corr-4d, increase in Isc mediated by both forskolin and genistein was observed, whereas Corr-4c increased genistein, but not forskolin, response (p<0.05 vs. vehicle control). The activity profile (across both cell lines) of Corr-4c and -4d resembles that of Corr-4a, and may provide a basis to discern distinct mechanisms induced by particular compound scaffolds or chemical subgroups (e.g. Corr-15Jf). Recent preclinical and clinical results using VX-770 suggest the primary airway cell model may predict in vivo efficacy. Complementary studies of the CFFT library in fully differentiated primary human bronchial epithelial monolayers are in progress, and preliminary results suggest correction in some (but not all) ∆F508/∆F508 donor cells. Comparison with cell lines may help determine the optimal model for future modulator screening. In aggregate, these findings highlight the importance of secondary studies in multiple cell models to evaluate compounds emerging from HTS programs. Difference in efficacy, activation profile, and activity across multiple cell models (i.e. spectrum of activity) may provide compound signatures that yield new insights regarding the mechanisms underlying correction of ∆F508 processing and gating, and could be used to construct optimized treatment paradigms for the ∆F508 mutation.

Supported by NIH and CFF. The validity of cystic fibrosis (CF) mice as a tool in pre-clinical testing of therapeutic strategies aimed at the correction of the ∆F508 folding, gating and cell surface stability defect depends on the assumption that human and mouse CFTR-∆F508, despite their limited sequence homology (78%) and differences in open probability (Po) and channel subconductance state, behave similarly in their response to pharmacological correctors and potentiators.

To verify this assumption, we have studied the responsiveness of mouse CFTR-∆F508 toward several classes of human CFTR-∆F508 correctors and potentiators in two assay systems: (i) CHO cells stably transfected with mouse CFTR-∆F508; (ii) freshly excised ileal mucosa from ∆F508/∆F508-Cftr-tm1Eur mice mounted in Ussing chambers.

All hCFTR-∆F508 correctors tested, including the Vertex compounds VRT-325 and -640, the histone deacetylase inhibitor DH1 (Balch lab), and the Verkman correctors 1a, 4a, 5c, 5i, 4c and 4d caused a moderate to strong dose-dependent rescue of mCFTR-∆F508, as evidenced by a 2-6 fold increase in forskolin+genistein-stimulated 125 Iodide efflux following 26h of incubation at 37 o C (CHO assay). EC50 values for rescue of mouse and human CFTR-∆F508 were similar. A subset of fast-acting correctors (VRT-640; VRT-325) was additionally capable of enhancing the forskolin+genistein-activated short-circuit current across ileal mucosa by 2-3 fold (up to 50-70% of the response in Cftr+/+ littermates) following 6h of ex vivo incubation in William's E Glutamax medium (37 o C). The maximal level of pharmacological correction reached was comparable to the effect of low-temperature incubation (26h, 26 o C) in both assays. Correction was associated with the appearance of mature band C CFTR protein on Western blots, and immunostainable CFTR in intestinal crypt cells.

In contrast, most hCFTR-∆F508 potentiators tested over a wide concentration range (0.1-100 mM), including NPPB-AM and the CFFT toolbox potentiators P1-P5 (VRT-532; PG-01; SF-03; UCCF-853; DF508act-02), P7 and P10, failed to potentiate forskolin-activated mCFTR-∆F508, both in low-temperature-rescued CHO cells and native intestinal epithelium. Only 3 of the CFFT compounds, P6 (genistein), P8 and P9, were able to potentiate mCFTR-∆F508 in both models, with similar potency and efficacy as hCFTR-∆F508 (2.5-6 fold enhancement of forskolin-induced 125 Iodide efflux).

These results indicate that mouse and human CFTR-∆F508 (1) respond to low temperature incubation and to small molecule correctors with a sim-ilar gain in surface expression and function, supporting the use of CF mouse models for in vivo tests of CFTR correctors; (2) respond differently to various classes of CFTR potentiators, in line with their pronounced differences in gating behaviour, and emphasizing the specificity of CFTR-potentiator interaction. Human/mouse CFTR chimera will be exploited in future studies to identify the domain(s) accounting for the differential response to the CFTR potentiators. Mapping of this channel gating region may facilitate the design of novel and more potent CFTR channel openers.

Bob Bridges, Rosalind Franklin University, and Cystic Fibrosis Foundation Therapeutics are acknowledged for providing us with the CFFT modulator library. In the development of cystic fibrosis (CF) transmembrane conductance regulator (CFTR) pharmacotherapy, new compounds that can correct processing defects of the CFTR protein (correctors) and/or improve the chloride (Cl -) channel activity of the mutated protein (potentiators) are continuously identified. Most promising compounds have been evaluated and optimised in vitro on cell cultures and animal models. However, their efficacy in human CF epithelia is unknown at the preclinical stage of development. Aim of this ongoing study therefore is to establish ex vivo ion transport measurements on human rectal biopsies as a new preclinical model for evaluation of potentiators and correctors.

We are testing the short-term (10min) Long-term incubation of rectal tissue with correctors could firstly demonstrate i) evidence for tissue vitality after 24h ex vivo incubation and ii) an improvement in CFTR-mediated Cltransport in F508del homozygous (mean ∆I sc forskolin+IBMX+carbachol corr4a: +5.6 µA/cm 2 t24/t0; VRT325: +4.6 µA/cm 2 t4/t0) but not compound heterozygous (F508del/G551D, F508del/R347P) CF patients and control. The effect of different potentiators on I sc is currently under investigation.

This first evidence of F508del-CFTR correction after long-term ex vivo incubation of human CF rectal tissue with small molecule compounds supports the important role of ICM as an outcome assay at the preclinical stage of CFTR pharmacotherapy development. Due to size of study cohort and unknown molecular corrector mechanism, the observed mutation-specific differences in drug effects (F508del homozygous vs. compound heterozygous) are too preliminary to discuss. Priorisation of compounds by ICM can be an essential step in the translation of CFTR-correcting/potentiating drugs into future clinical trials aiming to rescue the CF basic defect.

Supported by CFF, USA, and a financial grant from Christiane Herzog Stiftung and Mukoviszidose e.V., Germany.

Kim Chiaw, P. 1, 4 ; Huan, L. 1 ; Gagnon, S. 3 ; Ly, D. 2 ; Sweezey, N. 3, 5 ; Rotin, D. 2, 4 ; Deber, C.M. 1, 4 ; Bear, C.E. 1 Mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) lead to development of cystic fibrosis (CF). The major CF causing mutant, deltaF508-CFTR, results in a misfolded protein that fails to traffic out of the endoplasmic reticulum (ER). Introduction of second site mutations that disrupt a di-arginine (RXR)-based ER retention motif in the first nucleotide binding domain (NBD1) (R 553 XR 555 ) rescues the trafficking defect of deltaF508-CFTR. We hypothesize that if the R 553 XR 555 motif mediates retention of the native deltaF508-CFTR protein in situ, peptides that mimic this motif (CF-RXR) should antagonize mistrafficking mediated by aberrant exposure of the endogenous RXR motif. We generated CF-RXR peptides conjugated to the cell-penetrating peptide Tat (CPP-CF-RXR) to facilitate intracellular delivery and investigated their efficacy in rescuing the mistrafficking and dysfunction of deltaF508-CFTR. Using N-glycan processing, we demonstrate that CPP-CF-RXR is effective at increasing surface expression of deltaF508-CFTR. The increase in surface expression in the presence of CPP-CF-RXR was confirmed using fluorescent antibodies directed against a deltaF508-CFTR construct bearing an HA-epitope in the fourth extracellular loop, as monitored by FACS. Furthermore, CPP-CF-RXR improved deltaF508-CFTR channel function as assessed in cell lines by YFP quenching and Iodide efflux assays as well as in respiratory epithelial tissues obtained from CF patients assessed by Ussing Chamber assays. Taken together, our data suggests that abnormal accessibility of the RXR motif present in NBD1 is a key determinant of mistrafficking of the major CF causing mutant and furthermore addition of peptides mimicking this motif is capable of rescuing deltaF508-CFTR trafficking and function.

Funding support provided by BREATHE (CCFF and CIHR) and CIHR Strategic Training Programme in Structural Biology of Membrane Proteins Linked to Diseases. Several compounds that potentiate and/or correct mutant CFTR function are in active preclinical or early clinical development for the treatment of cystic fibrosis (CF). Previous studies in patients with mild CF genotypes (CFTR class IV or V mutations) indicated that low levels (10-20%) of normal CFTR function are sufficient to reduce disease severity and may therefore have significant therapeutic effects in CF. Therefore, highly sensitive and specific assays will be necessary to determine effects of potentiator and corrector compounds on CFTR function in clinical trials. Our previous studies demonstrated that intestinal current measurements (ICM) are a sensitive method to quantitate wild-type and mutant CFTR function in native rectal biopsy tissues (Hirtz et al. Gastroenterology 2004; 127:1085 -1095 . However, the sensitivity of this assay to monitor effects of potentiators and/or correctors remains unknown. The aim of this study was to evaluate the sensitivity of ICM to detect effects of a pharmacological potentiator on CFTR function in native non-CF and CF rectal tissues. To achieve this goal, we used the potentiator compound 1-EBIO, previously shown to improve CFTR-mediated Clsecretion in cultured cells by coordinate activation of luminal CFTR Clchannels and basolateral Ca 2+ -dependent and clotrimazole-sensitive K + channels (Devor et al. Am J Physiol. 1996; 271:L785-95) , and tested its effects in rectal biopsies obtained from CF patients carrying mild CF genotypes (class IV or V mutations; n = 12), severe CF genotypes (class I-III mutations; n = 27) and non-CF controls (n = 18). To determine effects of 1-EBIO on CFTR-mediated Clsecretion, rectal biopsies were mounted in perfused micro-Ussing chambers, pretreated with amiloride (10 µM) and indomethacine (10 µM), and sequentially exposed to IBMX/forskolin (100 µM/1µM), 1-EBIO (500 µM) and clotrimazole (30 µM) or CFTR inh-172 (20 µM). In non-CF rectal biopsies, cAMP-dependent activation (IBMX/forskolin) induced a large Clsecretory response (∆I sc-cAMP = -130.3 ± 19.2 µA/cm 2 ) that was further increased by 1-EBIO (∆I sc-EBIO = -50.2 ± 13.1 µA/cm 2 ; P < 0.001). 1-EBIO-induced I sc was significantly inhibited by clotrimazole and CFTR inh-172 . In rectal biopsies from CF patients with mild CFTR mutations, cAMP stimulation induced residual Clsecretory responses (∆I sc-cAMP = -40.7 ± 13.1 µA/cm 2 ) in the range of ~31% of non-CF controls. In this group, 1-EBIO further increased residual CFTR-mediated Clsecretion by ~33% (∆I sc-EBIO = -16.4 ± 5.1 µA/cm 2 ; P < 0.001). In contrast, cAMP activation and 1-EBIO failed to induce Clsecretion in rectal biopsies from CF patients with severe CFTR mutations. We conclude that (i) 1-EBIO potentiates CFTR-mediated Clsecretion by coactivation of luminal CFTR Clchannels and basolateral K + channels, and (ii) that ICM is a sensitive technique to monitor pharmacological activation of low levels of residual CFTR function in native rectal tissues. These results provide a proof of concept that ICM is a suitable biomarker for the assessment of potentiator and/or corrector effects on CFTR function in future clinical trials.

Supported by: Mukoviszidose e.V. Background: Clinical trial conduct in preschool children with cystic fibrosis (CF) has been hindered by lack of sensitive outcome measures transferable across multiple sites. We conducted a longitudinal, multicenter study of spirometry, inductance plethysmography (IP) and forced oscillometry (FO) in 3 to 6 year old children with CF and healthy controls, utilizing standardized equipment, rigorous site training, quality control, and an expert panel that reviewed all lung function data for acceptability.

Methods: Children with CF ages 36 to 60 months participated in up to 4 study visits ~6 months apart, plus a 2-week reliability visit. Healthy agematched controls had 1 study visit. Spirometry was performed at 8 sites; FO and IP were also performed at 5 of these sites. Z-scores for lung function measures were computed based on published equations. Mean z-scores from CF patients and controls were compared by linear regression (enrollment visit) and mixed effects models (repeated measures). Among CF patients, a mixed effects model was also used to evaluate the mean change in z-score with age.

Results: Ninety-seven CF patients and 87 controls were enrolled at 8 sites. Both groups had a mean age of 49.0 (SD 7.5) months at enrollment. Technically acceptable spirometry based on ATS guidelines was performed at 59% of 387 attempts; FO measurements were acceptable in 74% of 244 attempts. These proportions were similar for CF and controls. The mean spirometric z-scores for FEV0.5, FEV1 and FEF25-75 were significantly lower among the CF patients than the controls at the baseline visit, and declined significantly with age among the CF patients. The mean change/year in z-score (95% confidence intervals) was: -0.38, (-.55, -.22); -0.48 (-.73, -.24); and -0.53 (-.79, -.28), respectively. The mean FO z-scores from CF patients and controls were not significantly different, and did not change significantly with age among the CF patients. Results for IP will be presented.

Conclusions: The proportion of acceptable spirometry and FO measures was less in this multicenter study than in previously published single center studies. The spirometric measurements FEV0.5, FEV1 and FEF25-75 but not FO measurements were able to detect differences in average lung function between preschool CF patients and healthy controls, and the mean spirometric z-scores among the CF patients declined significantly with age. Spirometry may be more sensitive at detecting early disease than FO. The training and expertise gained by the sites through participation in this study has created a network primed to conduct future intervention studies involving preschool lung function testing. Objective: Pulmonary exacerbations, defined by an increase in respiratory symptoms and/or a decrease in pulmonary function, are a hallmark of cystic fibrosis (CF) lung disease. Management often requires hospitalization for IV antibiotics and increased frequency of airway clearance. In older children and adults, changes in pulmonary function testing are frequently used to determine response to therapy; however, little objective data is available on therapeutic response in infants. Our objective was to determine the change in lung function testing in a population of CF infants who underwent clinically indicated infant pulmonary function testing (PFTs) before and after hospitalization for management of pulmonary exacerbation.

Methods: This IRB-approved retrospective study was conducted using the infant PFT database at the University of North Carolina Chapel Hill. Subjects were included in analysis if they carried a diagnosis of CF and had undergone infant PFTs immediately prior to and following hospitalization for pulmonary exacerbation. PFTs were performed using the raised volume rapid thoracoabdominal compression technique and plethysmography to measure forced vital capacity (FVC), forced expiratory volume in 0.5 seconds (FEV 0.5 ), forced expiratory flows at 75%, 85%, and 25-75% of FVC (FEF 75 , FEF 85 , and FEF 25-75 , respectively), residual volume (RV), RV to total lung capacity (RV/TLC), and functional residual capacity (FRC).

Results: Final analysis included PFTs on 10 CF infants aged 36-166 weeks at the time of initial PFTs (mean 102 wks); 50% were male. Mean time from pre-antibiotic PFTs to post-antibiotic PFTs was 17 days. Initial PFTs showed evidence of substantial disease, with mean FVC, FEV 0.5 , FEF 75 , FEF 85 , and FEF 25-75 of 76. 1, 63.7. 30.5, 27, and 41 .2% predicted, respectively; means for FRC, RV and RV/TLC were 155.8, 186.2 and 184.1% predicted, respectively. Following antibiotics, there was a mean increase of 22, 33, 114, 132, and 86% above pre-treatment values for FVC (p=0.002), FEV 0.5 (p=0.0002), FEF 75 (p=0.01), FEF 85 (p=0.02), and FEF 25-75 (p=0.008), respectively. Lung volumes post-antibiotics also improved significantly, with mean decreases of 12, 15, and 18% in FRC (p=0.02), RV (p=0.03), and RV/TLC (p=0.007), respectively, from pre-treatment values. Initial PFTs showed evidence of airway obstruction (defined as FEV 0.5 <80% predicted and/or FEF 25-75 ≤70% predicted) in 90% of subjects; 90% had evidence of airtrapping (defined as RV/TLC and/or RV > 120% predicted). After hospitalization for IV antibiotics, only 40% of subjects persisted with evidence of obstruction, but 80% continued to have evidence of airtrapping.

Conclusions: There was a significant improvement in pulmonary function following IV antibiotics for pulmonary exacerbation in our population of CF infants who underwent PFTs pre-and post-hospitalization. The 31% increase in FEV 0.5 is particularly striking, suggesting dramatically reversible airways obstruction in these infants. Though there was a significant reduction in RV and RV/TLC % predicted values following antibiotics, these measures remained elevated in the majority of subjects. Persistently elevated RV and RV/TLC following therapies may reflect irreversible lung damage in peripheral airways and the progressive nature of CF lung disease. Rationale: Newborn screening (NBS) for cystic fibrosis (CF) is common, however, the lack of proven therapies for infants that delay the onset of lung disease indicate a need for randomized controlled trials that will require validated outcomes. Bronchiectasis is an important predictor of morbidity and mortality in CF and a significant clinical endpoint.

Objectives: To determine: (i) prevalence of bronchiectasis in young children with CF diagnosed following NBS; (ii) relations of bronchiectasis to pulmonary inflammation and infection; (iii) the persistence of CT changes.

Methods: Children were diagnosed with CF following NBS. Computed tomography (CT) and bronchoalveolar lavage (BAL) were performed during anaesthesia (n=96). CT scans were analysed for the presence and extent of abnormalities. BAL aspirate was cultured to identify possible viral, bacterial and fungal pathogens and inflammatory markers (neutrophils, neutrophil elastase) were quantified. From the same cohort, 181 pairs of CT scans obtained one year apart were examined to determine whether initially observed changes persisted.

Results: The prevalence of bronchiectasis was 22% and increased with age (p=0.001). Factors associated with bronchiectasis included, absolute neutrophil count (p=0.03), neutrophil elastase concentration, (p=0.001) and Pseudomonas aeruginosa infection (p=0.03). Of the 181 pairs of scans, bronchiectasis was present in 61 of the first scans and persisted in 45 (75%) to the next scan. Bronchiectasis was absent in 119 of the first scans but had developed in 65 of the pairs (55%) by the next CT.

Conclusions: Pulmonary abnormalities are common in infants and young children with CF and relate to neutrophilic inflammation and infection with Pseudomonas aeruginosa. Bronchiectasis observed on CT scans is persistent in the majority of cases. Current models of care for infants with CF fail to prevent respiratory sequelae. Bronchiectasis is a clinically relevant endpoint that could be used for intervention trials that commence soon after CF is diagnosed following NBS.

Presented on behalf of the Australian Respiratory Early Surveillance Team for Cystic Fibrosis (AREST CF -www.arestcf.org).

Zirbes, J.M.; Milla, C. Pediatrics, Stanford University, Palo Alto, CA, USA Methodology is available for the measurement of lung volumes and forced expiratory flows in infants comparable to older children and adults. However, these tests require sedation and are time-consuming. The multibreath washout (MBW) test involves the measurement of a tracer gas as it is either breathed in and/or out by a subject. MBW can be performed during regular tidal breathing in subjects of any age. Besides Functional Residual Capacity (FRC), one of the measurements that can be obtained is the Lung Clearance Index (LCI). LCI has advantages over spirometry in that it can be measured easily in all age groups and normative values are similar from early childhood to adulthood, which is not the case with most other PFTs. The recent availability of ultrasonic gas mass-molar ratio analyzers with fast response rates has permitted the development of devices that are more practical for the evaluation of LCI in infants. We wanted to evaluate the feasibility of conducting these measurements in infants with cystic fibrosis (CF) and compare their performance to other coventional measurements. The MBW procedure has been implemented at Stanford University with the use of a Exhalyzer ® D (Eco Medics AG, Switzerland). The tracer gas used for the measurements is 4% sulfur hexafluoride (SF6). Measurements were performed at the time of conventional infant PFT. Once the test was completed, the infant breathed through a face mask connected to a bypass flow arranged to provide room air at a flow that met the infant's inspiratory demand. The tracer gas was blended in to the bypass flow and its flow was controlled manually to start and end the gas wash-in. This was determined by visual inspection of the breath-by-breath time-concentration curve. A total of three maneuvers free of artifact were performed and the data saved for later analysis. We have at this time been able to test 9 infants with CF who underwent routine pulmonary evaluation. The testing procedure was well tolerated and completed without difficulty in all subjects. In 7 of these infants the measurement were performed under sedation and after completing RVRTC and plethysmography measurements. In the remaining 2 the measurements were performed without sedation. The mean (SD) LCI observed in this group was 10.7 ± 2.3, with the reported normal in healthy infants being <7.8. For the infants where plethysmography data was available, the FRC obtained by MBW correlated fairly well with the plethysmographic measurements (r=0.80). Of note is that although there was some correlation between LCI and RVRTC parameters that assess for the presence of airflow obstruction (r=0.46 for FEV0.5 and r=0.48 for FEF25-75), even infants with normal RVRTC parameters (greater than 85% predicted) had abnormal values for LCI. We conclude that MBW is a safe and feasible method for the assessment of pulmonary function in infants. With our methodology the LCI is easily estimated and identifies the presence of ventilation inhomogeneity among this group of infants with CF, as previously reported. Further, the LCI seems to be more sensitive to the presence of early airway disease in children than other conventional measurements of airway obstruction. LCI has good potential as a surrogate measure of airway obstruction in children with cystic fibrosis.

Ranganathan, S. 1,2 ; Schulzke, S. 5, 6 ; Nolan, G. 1,3 ; Logie, K. 3, 6 ; Murray, C. 3 ; Stick, S.M. 3, 6 ; Sly, P. 4, 3 ; Hall, G. 3 Structural changes such as bronchiectasis can be identified by computerised tomography (CT) in infants with cystic fibrosis (CF) (1) . In older children indices of ventilation distribution, such as lung clearance index (LCI) and mean dilution numbers (MDN), are considered to be sensitive markers of bronchiectasis (2) . Similar findings in infancy would enable early detection of lung disease without radiation. We assessed structural lung disease and its association with LCI and MDN obtained using multiple breath washout (MBW) in infants and young children with CF diagnosed by newborn screening (NBS).

Methods: Following sedation subjects underwent MBW with 5% sulphur hexafluoride and an ultrasonic flowmeter. LCI, first and second MDN were obtained. CT was done 1-2 days after MBW under intravenous anaesthesia using a three slice protocol at a lung inflation pressure of 25cm H 2 O and again at relaxed lung volume. The presence of bronchiectasis, airway wall thickening and air trapping were determined by a pediatric respiratory radiologist who was blinded to the results of lung function.

Results: Subjects were aged 2-27 months (n=46: 20 male). Structural changes were common (bronchiectasis n=14, bronchial wall thickening n=27, air trapping n=25). LCI was inversely correlated with age (r=-0.49;p<0.001: see figure) but was not increased (p>0.4) in infants with bronchiectasis (6.82 vs 6.95 without), bronchial wall thickening (6.83 vs 7.01 without) or air trapping (6.97 vs 6.84 without). Similarly, MDN were not increased in infants with structural changes.

Conclusions: Structural damage is common in infants and young children with CF diagnosed by NBS. Indices of ventilation distribution assessed by MBW were not associated with the presence of structural changes and are not useful as surrogates of early structural lung damage. Objective: Arikace™ is a sustained-release lipid formulation of amikacin for inhalation, being developed for the treatment of chronic P.aeruginosa (Pa) infection in CF patients. The primary objective of this study was to evaluate the safety, tolerability, pharmacokinetics, and efficacy of 28 day (d) exposure to once daily Arikace™.

Methods: This is a Phase 2 study conducted at 18 centers across the US. Subjects were stratified by FEV 1 % of predicted at baseline and randomized to receive either Arikace™ or placebo (1.5% NaCl) in a 2:1 ratio. Cohort 1 received 70 mg and Cohort 2 140 mg of active drug or placebo once daily for 28d by inhalation with eFlow® nebulizer system (PARI Pharma GmbH) and were then followed for 28d after completing study drug. Safety, pharmacokinetics, lung function, Pa sputum density, Quality of Life (CFQ-R) and rate of pulmonary exacerbation were evaluated weekly during the study period of 56d. Based on DSMB review of interim safety and efficacy data (in combination with final data from a similar parallel study in Europe), the study was amended to add Cohort 3, 560 mg once daily for 28d, and extended the evaluation for 56d after completion of study drug, providing a total study period of 84 days.

Results: A total of 46 subjects were enrolled into the study, and 21 adults in the first 2 cohorts have completed all aspects of the protocol. Baseline FEV 1 % pred. was 40-75% in 16 subjects and >75% in 5 subjects. Twenty-five subjects were enrolled in Cohort 3: 23 adults and 2 children ages 6-12 yr. Baseline FEV 1 % pred. was 40-75% in 19 subjects and >75% in 6 subjects. All subjects enrolled in Cohort 3 have completed 28d of study drug and to date 12 subjects have completed 84d of follow-up. The CFF DSMB reviewed interim safety and efficacy data from Cohort 3 after 15 subjects had completed 28d of treatment and recommended completion of the study without modification. Final data from the study will be presented at the NACFC.

Conclusion: Arikace™ given once daily for 28d at doses of 70 mg, 140 mg, and 560 mg was well tolerated in subjects with CF. Twenty-five subjects in Cohort 3 (560 mg) have completed treatment and are being followed for durability of response for 2 months off-treatment. Baseline characteristics and clinical parameters of the subjects enrolled in this randomized, placebocontrolled study of Arikace™ for inhalation may suggest a profile of subjects with CF who may benefit from this treatment. Cystic fibrosis (CF) is caused by mutations in the gene encoding CFTR, an ion channel involved in fluid regulation across epithelia. Sweat chloride is a biomarker used in the diagnosis of CF, with chloride levels ≥60 mmol/L indicative of disease. VX-770 is a CFTR potentiator under evaluation for the treatment of CF. As a systemic agent, it is hoped that VX-770 would improve CFTR function in the lungs and other organs, including the sweat gland. This randomized, double-blind, placebo-controlled study evaluated the safety and tolerability of VX-770 in adults with CF and a G551D-CFTR allele. Sweat chloride was also tested. Part 1 was a cross-over design that randomized subjects to placebo (n=4) or VX-770 q12h at either 25 and 75 mg (n=8) or 75 and 150 mg (n=8) doses in two 14-day periods. Part 2 included distinct subjects from Part 1 randomized to placebo (n=4), 150 mg VX-770 (n=8), or 250 mg VX-770 (n=7) q12h for 28 days. Throughout the study VX-770 was generally well-tolerated and no drug-related serious adverse events occurred. In Part 1, mean±SD baseline sweat chloride for the study population was 101.4±12.9 mmol/L. At Day 14, mean chloride was near or below the diagnostic threshold for CF in the VX-770 75 mg (60.9±21.8 mmol/L) and 150 mg (47.5±14.8 mmol/L) groups. Changes from baseline in the VX-770 groups were significant (P<0.001 within-group and vs placebo). In Part 2, the median (range) sweat chloride concentration at baseline was 98.6 (80.5-117.0) mmol/L. At Day 28, sweat chloride levels were 52.0 (28.0-84.0) and 64.0 (39.0-77.5) mmol/L in the VX-770 150 and 250 mg groups, respectively, versus 102.5 (100.0-107.0) mmol/L for placebo. The changes from baseline in the VX-770 groups were significant (P<0.05 within-group and vs placebo) at the first treatment assessment (Day 3) and at treatment end (Day 28). Combining Parts 1 and 2 at Day 14, a total of 6 (75.0%), 11 (84.6%), 13 (92.9%), and 4 (57.1%) subjects in the VX-770 25, 75, 150, and 250 mg dose groups, respectively, were responders by the pre-specified criterion of a change from baseline of ≥20 mmol/L. No placebo-treated subjects were responders. Sweat sodium changes with VX-770 paralleled sweat chloride. Serum aldosterone was measured in Part 1 and did not change significantly. Correlations between change from baseline in FEV1 and sweat chloride in treatment groups were generally weak and not conclusive, which may be due to the small sample size. In conclusion, a rapid significant improvement in sweat chloride followed administration of VX-770 that was near or below the diagnostic threshold for CF, suggesting that further evaluation of VX-770 in larger studies is warranted. Background: This open-label study included 274 patients with CF (age > 6 years with pulmonary Pseudomonas aeruginosa (PA) and FEV1 > 25% and < 75% predicted) from 71 study centers (US, Canada, Australia, New Zealand) who participated in one of two previous double-blind studies of AZLI 75 mg (dosed twice [BID] or three times [TID] daily). Patients received AZLI (as assigned in the previous studies) for nine 28 day courses, separated by 28 day off treatment intervals.

Results: Clinical efficacy was maintained over the nine courses of AZLI. Patients receiving AZLI TID showed greater improvement in FEV1, and CFQ-R Respiratory Symptoms Score, and showed a greater reduction in log 10 PA CFUs after each of the nine treatment courses. Median time to hospitalization was greater for patients receiving AZLI TID. The most common treatment-emergent AEs reported with an incidence rate ≥ 50% were those related to respiratory conditions consistent with CF. The two most commonly reported treatment-emergent AEs were cough and productive cough, followed by exercise tolerance decreased and respiratory tract congestion. Common AEs were reported more frequently during off-AZLI periods than on-AZLI. There was no evidence that the incidence of any common AE increased with repeated exposure.

Conclusion: AZLI was generally well-tolerated over nine courses of administration (18 months), and TID-treated patients showed greater improvement than BID-treated patients in disease-related endpoints.

Supported by Gilead Sciences, Inc. Modulation of Pseudomonas aeruginosa (Pa) virulence factors was suggested as mechanism for azithromycin (AZM) beneficial effects in cystic fibrosis (CF) patients. Patients with cystic fibrosis are particularly susceptible to chronic Pa infection in the airways. Our work was aimed to study the regulation of proteins released by Pa strains after AZM treatment.

Looking for secreted virulence factors active on lung epithelium of CF patients we focused on the induction of pro-inflammatory markers and measured the expression of IL-8 gene in 16HBE14o-AS3 CF epithelial airway cell line in response to conditioned medium (CM) derived from Pa strains. P. aeruginosa clinical isolates and PAO1 were grown at stationary phase under aerobiosis in the presence or absence of AZM. We demonstrated that conditioned medium from the clinical strain AA2, unlike CM from the laboratory strain PAO1, induced a statistically significant increase of about 4 times of IL-8 mRNA, in CF airway epithelial cells. This induction was reduced about 20% when AA2 was grown in the presence of AZM, suggesting that this macrolide reduces Pa pathogenicity. In the attempt to gain information on the identity of the molecules released by Pa strains before and after treatment with AZM and to identify candidate molecules involved in Pa virulence, we applied a recent proteomic approach (2DC-MS/MS). Two-dimensional capillary chromatography -tandem mass spectrometry (also named MudPIT) identified polypeptides released by PAO1 and AA2 in the absence and presence of drug. Seven proteases were released from AA2 while only one was detected in PAO1. AZM appeared to down-regulate their release in AA2 strain. This result has been confirmed by zymography.

Focusing on released Pa factors is critical for identification of molecules involved in bacterial virulence, lung damage, inflammation and drug resistance. This approach could identify potential targets for pharmacological intervention and provide candidates for cellular pathways leading to better understanding of CF pathogenesis.

Supported by Italian Cystic Fibrosis Research Foundation(FFCgrant#17/2006); Comitato Vicenza-Associazione Veneta Lotta contro la Fibrosi Cistica. Recombinant adeno-associated virus (rAAV) represents a promising potential gene therapy vector for cystic fibrosis (CF). However, the most commonly-studied rAAV serotype (rAAV2) is inefficient at transducing human airway epithelial (HAE) cells from the apical membrane. Interestingly, rAAV transduction from the basolateral membrane is highly efficient. In contrast, rAAV1 is efficient at transducing HAE with no polarity bias. Differences in intracellular trafficking of rAAV virions have been thought to underlie the transduction differences observed with these two serotypes. We hypothesized that rAAV1 and rAAV2 utilize similar intracellular trafficking pathways from the basolateral membrane of polarized HAE, resulting in efficient transduction. However, we surmise that these two serotypes utilize distinct apical endosomal trafficking pathways, leading to efficient nuclear import and transduction in the case of rAAV1, but resulting in poor transduction with rAAV2 due to routing of the virus to a compartment that inefficiently processes the virion. In order to test these hypotheses, we conducted transduction experiments using rAAV particles labeled with fluorescent dyes. Following apical infection of polarized HAE cultures with labeled rAAV1 or rAAV2, rAAV1 was distributed throughout the cytoplasm of epithelial cells while rAAV2 was sequestered in the apical cytoplasm. These findings support inefficient intracellular trafficking of rAAV2 as compared to rAAV1. In addition, we utilized co-infections with two sets of virions labeled with different fluorophores to directly compare the apical and basolateral trafficking of rAAV1 or rAAV2, and also to directly compare rAAV1 and rAAV2 trafficking from either the apical or basolateral membranes in the same cells. Greater co-localization was observed between apically and basolaterally-applied rAAV1 than for rAAV2, indicating that rAAV1 uses similar intracellular trafficking pathways for apical and basolateral infection, but rAAV2 does not. In addition, we observed little co-localization between rAAV1 and rAAV2 applied simultaneously from the apical membrane, suggesting that these viruses use distinct apical trafficking pathways. In contrast, basolateral co-infection resulted in more significant co-localization between rAAV1 and rAAV2, demonstrating that these viruses utilize similar trafficking pathways following basolateral infection. We also analyzed the requirements of the cellular proteins Rac1 and Rab5, which are involved in vesicular trafficking, on rAAV apical and basolateral transduction using dominant-negative constructs. We found that rAAV1 and rAAV2 both require these proteins for efficient transduction from the apical and basolateral membranes, but to different degrees, highlighting the distinct biology of intracellular trafficking in HAE cells between the different serotypes. The results of this study have shown that intracellular trafficking of rAAV is directly influenced by the viral capsid sequence. An understanding of the basic biology of rAAV trafficking in HAE cells will be essential for determining the particular characteristics needed for a successful gene therapy vector for CF. The design of safe and efficient vectors for gene delivery is a current challenge for cystic fibrosis (CF) gene therapy. Viral-based systems are for now, one of the most efficient methods to deliver genes into the cells. However, immunogenic and oncogenic complications are two of the main disadvantages of this transfection approach. As an alternative, synthetic carriers have been developed, notably based on cationic lipids. We have previously demonstrated the efficiency of our original vector, especially in vivo. However, to enhance cationic lipid transfection ability, their design and formulation have to be adapted to overcome the extra-and intracellular-barriers.

Here, our objective was to evaluate the in vivo transfection capacity of three cationic carriers (KLN47, EP8e, BSV-18) that differ by the nature of their hydrophobic domain, respectively oleyl, linoleyl and phytanyl chains. Formulated in NaCl 0.9%, they are associated with 50 µg of pDNA encoding Luciferase. Then, lipoplexes are injected in the tail vein of mice and Luc expression was analyzed up to 72 h by in vivo bioluminometer (NightOwl II, Berthold). We first observed that Luc expression was exclusively located into the lungs. Secondly, the highest transgene expression was obtained 24 h after transfection and progressively decreased up to three days. Thirdly, EP8e and BSV-18 were much more efficient than KLN47. Then, contrary to KLN47, both new vectors did not induce any mice death and their lower toxicity was confirmed by ALAT/ASAT measurements. Moreover, if they induced a very weak inflammation, no immunogenicity stimulation was measured. The structure-activity relationship is crucial for a gain of effectiveness and innocuity. The majority of cystic fibrosis (CF) patients acquire chronic Pseudomonas aeruginosa lung infection. Treatment of the chronic biofilm infection is difficult due to tolerance and development of resistance to antibiotics. Therefore, adjunctive treatment possibilities are pivotal. The course of the chronic P. aeruginosa lung infection is characterized by development of a pronounced antibody production and a T-helper 2 (Th2)-dominated immune response, which correlates to a poor prognosis. In contrast, a more Th1-dominated response in CF has been shown to correlate with a better lung function. Since the serum concentrations of GM-CSF in CF correlated with a more Th1-dominated response, we speculated whether GM-CSF treatment could induce a Th1 response in susceptible Th2-reacting BALB/c mice with chronic P. aeruginosa lung infection. Dendritic cells (DC) would presumably be mediating such effect.

Eighteen female BALB/c mice were infected in the lungs with P. aeruginosa embedded in seaweed alginate to resemble the chronic biofilm lung infection in CF. The mice were divided into 3 treatment groups with 6 mice in each group: a high dose group (0.2 µg/day), a low dose group (0.02 µg/day) and placebo treated controls. Mice were treated from day 3 until day 6, and sacrificed day 7. IFN-γ (Th1 marker), IL-4 and IL5 (Th2 markers) released by isolated lung cells, and the fractions and subsets of pulmonary dendritic cells were estimated by ELISA and FACS analysis respectively.

The IFN-γ/IL-4 and the IFN-γ/IL-5 ratio increased significantly (p<0.02) in the high dose treatment group indicating induction of a Th1-dominated response. Using magnetic cell sorting revealed that both CD4+ (T-helper cells) and CD8+ (cytotoxic T cells) cells contributed to the IFN-γ production, whereas the IL-4 production originated from CD4+ cells exclusively. In addition, the IFN-γ/IL-4 ratio doubled in CD4+ cells in the high dose treatment group. CD8α and CD11b surface expression was used to distinguish between different subsets of DCs. The fractions of double positive CD8α/CD11b DCs (Langerhans cell (LC) -derived DC) increased significantly in the high treatment group (p<0.04), whereas the fractions of CD8α or CD11b single positive DCs did not differ between the 3 groups.

The present study indicates a Th1-inducing role for GM-CSF in patients with chronic P. aeruginosa lung infections. In addition, the GM-CSFinduced modulation of the Th1/Th2 balance is probably mediated through the pulmonary LC-derived DCs. Cystic fibrosis (CF) is caused by mutations in the CF transmembrane conductance regulator (CFTR) gene. The most common mutation, ∆F508-CFTR, is a temperature-sensitive, trafficking mutant with reduced chloride transport and exaggerated immune response. The ∆F508-CFTR is misfolded, ubiquitinated, and prematurely degraded by proteasome mediateddegradation. We recently demonstrated that selective inhibition of proteasomal pathway by the drug bortezomib (pyrazylcarbonyl-Phe-Leu-boronate, a.k.a. Velcade or PS-341) ameliorates inflammatory pathophysiology in CF cells. This proteasomal drug is an extremely potent, stable, reversible and selective inhibitor of chymotryptic threonine protease activity. A main concern in considering the proteasome as a therapeutic target is the theoretical risk that multiple processes may be affected by proteasome inhibitors. The risk of global suppression of proteasomal degradation may outweigh the favorable anti-inflammatory effect we observed. To overcome this challenge, we developed a nano-based approach that uses drug loaded biodegradable nanoparticle (PLGA-PEG-Bortezomib) to provide controlled drug delivery at lower doses, which demonstrates the promise to rescue the CF lung disease in murine model. The in vitro release kinetics of drug from nanoparticle was quantified by proteasomal activity assay from days 1-7 that showed slow drug release from day 2-7 with maximum inhibition at day 7. For in vivo release kinetics, these drug-loaded nanoparticles were fluorescently labeled, and administered to wild type (WT) and CF knockout (CFKO) mice by intranasal route. Whole-body optical imaging of the treated live animals demonstrates efficient delivery of particles to murine lungs, 2 hrs post treatment, followed by biodegradation and release over time. The efficacy of drug release in WT and CFKO mice lungs was determined by quantifying the changes in proteasomal activity (< 2 fold) and ability to rescue the Pseudomonas aeruginosa LPS induced inflammation, 3-5 day post treatment. In conclusion, we have developed a novel drug delivery system that has been engineered to bypass the mucus barrier and provide sustained drug delivery in CF lungs. We are also developing methods to encapsulate selected known CF correctors, potentiators and antimicrobials, in PLGA-PEG based nanoparticles to develop this nanosystem as a therapeutic delivery vehicle for variety of CF drugs. E-mail: nvij1@jhmi.edu Support: Accelerated Translational Incubator Program (NIH CTSA UL RR 025005). The finding that the cystic fibrosis transmembrane conductance regulator (CFTR) is present in human red blood cells (RBC) opens new perspectives for blood-based diagnosis of cystic fibrosis (CF). We developed such a blood-based CF test in which the trivalent cation Gadolinium (Gd 3+ ) induces hemolysis of RBC from healthy donors but fails to hemolyse RBC from CF patients (Stumpf et al. Physiological concept for a blood based CFTR test. Cell Physiol Biochem 2006; 17:29-36) . Actually, the selectivity and sensitivity of this method will be tested in an ongoing multicentric clinical trial.

During the last years several substances were developed which are supposed to be beneficial CF therapeutics (potentiators, rescuing compounds). Rescuing compounds force misfolded CFTR (∆F508) into the plasma membrane while CFTR potentiators increase the chloride conductance of CFTR (G551D, rescued ∆F508). Especially the potentiators showed promising results and some of them are actually tested in clinical trials.

We tested if CFTR potentiators are able to enhance gadolinium induced hemolysis of human RBC. The motivation behind this study was to check if the CF blood test could be used as a fast readout parameter for the effect of CFTR potentiators. RBC are due to their high renewal rate presumably one of the first responding cells after systemic application of potentiators. The effect of the following potentiators on the gadolinium induced hemolysis of RBC from healthy donors were tested: VRT-532, PG-01, SF-03, UCCF-853, δF508act-02, NSOO4, 4-(4-oxo-4H-benzo[h]chromen-2-yl)-pyridinium, 3-But-3-ynyl-5-methoxy-1-phenyl-1H-pyrazole-4-carbaldehyde and 3-(2-Benzyloxy-phenyl)-5-chloromethyl-isoxazole. We also tested the CFTR specific inhibitors CFTRinh172, GlyH-101 and Blocker 5ab.

Without gadolinium, none of these substances were able to induce hemolysis in RBC. In combination with gadolinium the potentiator UCCF-853 increases the hemolysis strongly while VRT-532 and NSOO4 significantly reduced the gadolinium induced hemolysis. Also the CFTR specific blocker GlyH-101 was able to clearly reduce the gadolinium induced hemolysis. All other substances showed no effect.

These data show that UCCF-853, VRT-532, NSOO4 and GlyH-101 are not exclusively affecting the CFTR chloride conductance because the mechanism of the gadolinium induced hemolysis depends not exclusively on CFTR chloride conductance but also on the regulatory functions of CFTR. Therefore the obtained results are not so surprising particularly with regard to the fact that these experiments were performed (so far) only on RBC from healthy donors. Additionally, the RBC were incubated in vitro with the tested substances. Presumably we will observe effects of potentiators on the gadolinium induced hemolysis with RBC derived from CF patients and more likely after systemic administration of potentiators to CF-Patients.

Together, we state that further experiments with RBC from CF patients treated with potentiators are necessary to clarify the question, if the CF blood test is applicable to quantify the effect of potentiator CF therapeutics. Introduction: Chronic sinusitis is a very common disease in CF significantly impairing quality of life. There is also increasing evidence that microbes found in the upper respiratory tract of CF patients are the same as those found in the lower airways of CF patients. While aerosolized drug delivery to the lungs is an established treatment route, drug delivery systems to the paranasal cavities are not yet established. However, due to the unmet need of medical treatment in chronic sinusitis, a number of companies are selling sinonasal nebulizers and claim to deliver medications to the sinuses, but these claims are not supported by valid in vitro or in vivo studies.

Objectives: This study was carried out to compare sinonasal nebulizers including a novel prototype device (VibrENT) generating a fine aerosol mist via a pulsation aerosol principle utilizing a nasal cast model.

Methods: Four commercially available sinonasal nebulizers and a VibrENT prototype were tested according to their instructions of use. Nebulization efficiency was measured using a nasal cast model (Moeller, 2008, Rhinology) linked to a breath simulator mimicking nasal breathing. The nebulizers were filled with a novel levofloxacin formulation and nebulized for 8 min each. At the end of nebulization, the nasal cast was disassembled and drug from the paranasal cavities including ostia, the nasal cavity, as well as the inspiratory filter was assayed by HPLC.

Results: The VibrENT prototype delivered 1.0% of the loaded dose to the paranasal cavities and this deposition fraction increased to 19% mimicking a closed soft palate. All other devices tested delivered less than 0.06% of the charged drug to the paranasal cavities. Nasal cavity deposition was 3% for the SinuNeb, 12% for the Aeroneb Go and 50% for the VibrENT. In-vitro lung deposition was 2% for the Atomisor AMSA, 8% for the VibrENT and 12% for the Aeroneb Go.

Conclusions: Current sinunasal drug delivery systems do not provide adequate sinonasal drug deposition using a human nasal cast and may explain the poor therapeutic efficacy of current nasal nebulizers. The VibrENT prototype showed promising in vitro results but clinical studies are needed to validate in vitro results.

Sinus and nasal cavity deposition: µg levofloxacin found in all six sinus cavities including ostia and nasal cavity after eight minutes administration time. Chronic colonization with Pseudomonas aeruginosa and the resulting progressive loss of lung function ultimately leads to respiratory failure in many patients with cystic fibrosis (CF). The P. aeruginosa virulence factor hemolytic phospholipase C (PlcH) is an extracellular protein able to cleave the phosphorylcholine head group from phosphatidylcholine, a major component of lung surfactant, resulting in surfactant dysfunction. Surfactant dysfunction is hypothesized to be one of many factors resulting in poor lung function in P. aeruginosa colonized patients. We presented data at the previous NACFC meeting showing the significant impact of PlcH on host lung physiology during infection which suggested that PlcH inhibition may be a possible anti-pseudomonal therapy. The ether-linked phosphocholine compound miltefosine is a promising inhibitor of PlcHR enzyme function. The kinetic analysis of miltefosine inhibition of purified PlcH enzymatic activity will be presented. To test the efficacy of miltefosine as a potential therapeutic agent, we used our mouse model of P. aeruginosa PAO1 lung infection to determine the effects of miltefosine on infection progression and lung physiology as determined using a small animal ventilator (flexiVent, SCIREQ). Administration of miltefosine via intraperitoneal injection did not lead to differences in viable bacterial counts from the lungs and did not alter the number of immune cells in the airway lumen. However, measurements of lung mechanics, including pressure-volume loops and impedance mechanics, demonstrate significant protection of the surfactant system and resulting lung physiology. We are conducting dose titration studies of miltefosine to assess the optimal dose and delivery method to prevent lung damage by P. aeruginosa. These data suggest that, by limiting surfactant dysfunction, miltefosine may be an effective addition to the therapeutic regimen against P. aeruginosa . Ca 2+ activated Clchannels (CaCC) are co-expressed with CFTR in airway surface epithelia and up-regulated in cystic fibrosis (CF). The presence and functional properties of CaCC make it a possible therapeutic target to compensate for the deficiency in Clsecretion in CF epithelia. CaCC is activated by an increase in cytosolic Ca 2+ . The intracellular Ca 2+ increase not only activates epithelial CaCCs, but also inhibits epithelial Na + hyperabsorption, which may also beneficial in CF.

Our previous study had shown that spiperone, a known antipsychotic drug, activates CaCCs and stimulates Clsecretion in polarized human non-CF and CF airway epithelial cell monolayers in vitro, and in CFTR knockout mice in vivo. Spiperone activates CaCC not by acting in its well-known role as an antagonist of either 5-HT2 or D2 receptors, but through a protein tyrosine kinase-coupled phospholipase C-dependent pathway. Herein, we performed a mass spectrometry analysis and identified 32 proteins that showed significant differences in their expression level between spiperone treated and non-treated CF airway epithelial cells. Four of those proteins are involved in the tyrosine-phosphorylation pathway and three of them are expressed in our airway cells. Among these three proteins, proline-rich tyrosine kinase 2 (PYK2) has been shown to be implicated in PLC activation and increased release of Ca 2+ in endothelial cells. PYK2 is a cytoplasmic tyrosine kinase which is concentrated at the focal adhesion. We also found that the inhibition of PYK2 notably reduced the ability of spiperone to increase intracellular Ca 2+ .

In conclusion, we have identified the tyrosine kinase, PYK2, which plays a crucial role in spiperone's ability to increase intracellular Ca 2+ . This novel kinase may also be a potential therapeutic target for a new therapy for CF. Objectives: To evaluate the safety and tolerability of inhaled NO over a 2 day period in clinically stable subjects with CF and to investigate its short term effects on biomarkers of inflammation and infection.

Methods: In this randomized, double-blind, placebo-controlled, single center trial, 18 CF subjects (13M:5F; age 17±3 years; FEV1 94±15% predicted) were enrolled sequentially into a low-dose (20 ppm) NO cohort (6 active:3 placebo) followed by a high-dose (40 ppm) NO cohort (6 active:3 placebo). The NO (INOvent®, INO Therapeutics, Inc., Datex-Ohmeda) or placebo (nitrogen) was administered via nasal cannula over a 44 hour period. Safety was assessed by adverse event reporting, oxygen saturation monitoring, venous methemoglobin (metHgb) levels, spirometry, and echocardiography. Induced sputum measures of inflammation (total cell counts, % neutrophils, free neutrophil elastase activity, IL-8, nitrites, nitrates, 8-isoprostanes) and infection (bacterial colony counts) and breath condensate measures of inflammation (pH, nitrites, nitrates, 8-isoprostanes) were collected at baseline and 48 hours. An institutional Data Safety Monitoring Board performed a safety review after the initial low dose cohort.

Results: Inhaled NO was well tolerated by all study subjects. No subject experienced any of the following predefined adverse events: sustained decline in oxygen saturation, decline in FEV1>10%, or increase in metHgb>3%. One subject who was randomized to active drug in the lowdose cohort was removed from subsequent analysis because both NO and placebo gas were inadvertently given to that subject. The only significant change among the biomarkers of inflammation was an increase in sputum 8isoprostanes in the low dose group. A significant reduction in S. aureus colony counts and a near significant reduction in P. aeruginosa colony counts were observed in the high-dose cohort relative to placebo (p=0.03 and 0.06, respectively). Most measures of inflammation and infection in induced sputum from the placebo subjects demonstrated good reproducibility with low variability.

Conclusions: Inhaled NO by nasal cannula over 2 days was safe and well tolerated by clinically stable older CF children and young adults with mild lung disease. NO inhalation at 40 ppm demonstrated antimicrobial effects. These preliminary findings warrant further, longer term studies. Augmentation of airway NO by nasal inhalation is a promising antimicrobial approach for CF subjects during acute pulmonary exacerbations or even as chronic therapy in individuals with advanced lung disease. is TOBI ® delivered with the PARI LC PLUS ® nebulizer. Major disadvantages of this concept are long inhalation times up to 20 min leading to poor patient compliance. The eFlow®, a highly efficient electronic nebulizer, can reduce nebulisation time providing at the same time an equivalent tobramycin lung deposition, yielding an advantage for the patients.

Methods: Plasma and sputum tobramycin levels were assessed in a phase Ib, multi-center study (10 centers) in a randomized, parallel group design to compare the safety and pharmacokinetics of inhaled tobramycin. CF patients either inhaled the established product TOBI ® (300 mg/5 ml, delivered by PARI LC PLUS ® ) or Tobramycin PARI (=T100, 150 mg/1.5 ml, delivered by an investigational eFlow ® ) twice daily for 28 days. The goal of the study was to prove that equivalent or lower tobramycin plasma levels can be achieved. Seventy-eight CF-patients, including 42 children (8-17 yrs) and 36 adults (18-42 yrs) with a mean FEV1 % pred of 74.5 ± 20.3 and a mean age of 20.3 ± 8.9 years completed the study.

Results: Equivalent or higher sputum concentrations of tobramycin were achieved after 7 days of inhalation: T100 = 2.6 mg/ml (CI90% 2.0-3.2), TOBI = 2.3 mg/ml (CI90% 1.7-2.9), whereas plasma levels were the same or lower: Cmax of T100 = 1.29 mg/ml (CI90% 1.05-1.53), Cmax of TOBI = 1.65 (CI90% 1.41-1.89). The patients reported less side effects for T100 compared to TOBI indicating comparable or better tolerability, and importantly, the delivery time with the eFlow ® was 3 to 4 times faster with mean nebulisation times for T100 = 4.6 ± 1.2 min and TOBI= 16.1 ± 3.5 min.

Conclusions: This study proves the concept of a safe and much more rapid delivery of inhaled tobramycin to the lungs of patients with cystic fibrosis with an eFlow ® investigational nebulizer. Data also allow the assumption that drug adherence may be improved. Based on these potential benefits an orphan drug designation was granted in the EU. These aspects and therapeutic equivalence, i.e. improvement of FEV1 and an improvement of quality of life will be tested in a phase 3 study.

We thank the involved centres in Rabka, Gdansk and Warszawa, Poland and in Munich, Frankfurt, Berlin-Buch, Hannover, Erlangen and Essen-Haidhausen, Germany. Previously, we characterized the anti-inflammatory effect of synthetic triterpenoids in CF. We find that two of these compounds, CDDO and CDDO-Me, exhibit potent anti-inflammatory effects in a number of models of CF including: 9HTEo-(pCEP compared pCEPR (CF)) and 16HBEo-(sense compared antisense (CF)) cells, human primary tracheal epithelial cells (HPTE, treated with vehicle vs. 20µM CFTR inh -172 (CF)), and R117H mutant mice vs. normal littermates. Inflammation in our models is excessive compared with normal controls. In stimulated CF cell models CDDO decreases inflammatory cytokine (IL-8 and IL-6) production by 23-57% from non-treated cells, correcting excessive inflammation. A similar effect was observed for CDDO-Me in HPTE inhibited with the CFTR inh -172. In mice stimulated with LPS, cytokine production (IL-8, IL-6, TNFα, and IL-1β) was decreased to background levels when animals received intratracheal administration of CDDO prior to stimulation. No adverse effects were observed for any of the CDDO or CDDO-Me doses used (max in cells: 300 nM, max in animals: 5mg/kg). Taken together these data are compelling evidence for the potential utility of these compounds in CF. Therefore, we used 2 approaches to study the mechanism of action of these compounds. First, we treated HPTE cells and mutant R117H mice with 300 nM or 5 mg/kg CDDO, respectively. Three days later, we stimulated CFTR inh -172 inhibited HPTE cells with 10 ng/mL TNFα/IL-1β and CF mice with LPS. Controls received vehicle (25µL DMSO). Whole cell protein was prepared from HPTE cells or mouse lungs, separated by 2D gels, and compared to controls by densitometry. Differential expression of 103 proteins by >2 fold in HPTE cells treated with CDDO vs. vehicle was observed. Of these, 87 were concordant with proteins found to be different in CF mouse lungs treated with CDDO vs. vehicle. GeneGo TM pathway analysis revealed that CDDO activated stress responses, especially the Nrf2 regulated antioxidant/anti-electrophile response element, and improved the ability of cells to respond to stimulation. For our second approach, we conducted metabolomic screens of the effect of CDDO treatment on stimulated HPTE cells inhibited with CFTR inh -172 vs. controls. HPTE from 6 different individuals were cultured at an air liquid interface and allowed to form tight junctions. Metabolomic analyses revealed CDDO effects that were concordant with those observed in proteomic analyses, with enhancement of stress responses resulting in the normalization of stress responses. These data were consistent with increased activation of Nrf2 and the enhanced activation of the antioxidant/anti-electrophile response element. We conclude that CDDO and CDDO-Me correct cytokine production in CF models to normal levels. Normal responses are not diminished and no cell toxicity is observed for maximum doses. Proteomic and metabolomic screens of the effect of these compounds are consistent with a mechanism of action that involves the activation of Nrf2 and the antioxidant/anti-electrophile response element, which significantly corrects aberrant inflammatory responses in CF models.

Supported by the NIH, the CFF, and Reata Pharmaceuticals Inc. Background: Cystic fibrosis (CF) is the most common life limiting autosomal recessive disease and is due to mutations in CFTR gene. CF mutations cause a massive pro-inflammatory phenotype in the lung arising from aberrant expression of inflammatory genes. Recently microRNAs (miRNAs), a class of small endogenous RNA molecules, have emerged as important targets in the frontier of biomedical research. miRNAs regulate gene expression by targeting mRNAs for degradation and/or translational repression. The dysregulation of specific miRNAs has been demonstrated in a variety of diseases in humans including cancer, heart disease and diabetes. However, to date there has been no study targeted towards examining the role of miRNAs in CF. The present study is targeted to investigate how aberrant expression of miRNAs in CF lung epithelial cells leads to the disease phenotype.

Hypothesis: Our hypothesis is that microRNAs play a major role in causing aberrant expression of pro-inflammatory genes in the cystic fibrosis lung.

Methods: miRNA expression profiling of CF IB3-1 cells and AAV-CFTR repaired IB3-1/S9 cells was carried out using the TaqMan microRNA low density arrays comprising 365 different miRNAs (Applied Biosystems). The target genes for the mis-expressed miRNAs in CF cells were identified using three common miRNA databases, including TargetScan, PicTar and miRBase (Sanger's miRNA database). Functional pathway analysis of miR-NAs was carried out using both Ingenuity Pathway Analysis (IPA) and MetaCore (GeneGo) Analysis. The mRNA levels corresponding to miRNA expressions for CF IB3-1 and CFTR-repaired IB3-1/S9 cells were analyzed using ILLUMINA bead chip HumanRef-8 v2.0 arrays.

Results: miRNA profiling of CF IB3-1 cells as well as control IB3-S9 cells revealed that there are 22 miRNAs which exhibit altered expression levels in IB3-1 cells. Of these, 18 miRNAs are upregulated and 4 are downregulated in IB3-1 compared to IB3-1/S9 cells. Functional pathway analyses identified signaling pathways, including TLR4 and TGFβ signaling, relevant to the CF disease status. The corresponding mRNA expression in these cells further corroborates the miRNA expression data.

Conclusions: Our findings indicate that aberrant expression of miRNAs play a significant role in the inflammatory phenotype manifest in CF. Insight into the mechanisms associated with miRNA-mediated regulation of inflammation in CF will lead to novel miRNA-based therapy for CF. DNA nanoparticles (DNA NPs), a type of non-viral gene therapy, are formulated by condensing DNA plasmids with a polycation carrier that consists of 30 lysines coupled to a linear polyethylene glycol (PEG) through an N-terminal cysteine (CK 30 ). The ability of DNA NP to transfect airway epithelium in vivo and its desirable safety profile in human have made this gene transfer agent a promising candidate for novel therapeutic applications to cystic fibrosis (CF).

To further enhance the gene transfer efficiency of DNA NPs, we introduced a reducible disulfide linkage to the polycation carrier to aid the unpacking of DNA NPs and transgene expression. We previously found that in vitro gene transfer mediated by DNA NP prepared from PEGylated CK 30 with a disulfide bond (SS) linking PEG and CK 30 (SS-DNA NP) was over 20-fold more efficient than that mediated by DNA NP prepared with a thioether bond (CS) as the linkage (CS-DNA NP).

To study the mechanism of this enhanced in vitro gene transfer, we first examined the cellular uptake of the two DNA NPs fluorescently labeled by YOYO-1. Quantitative measurement by flow cytometry revealed higher cellular uptake of SS-DNA NP compared to CS-DNA NP, suggesting an extracellular mechanism is involved. DTNB, a cell impermeable free sulfhydryl blocker, inhibited in vitro gene transfer by SS-DNA NP in dose-dependent manner, indicating that this mechanism is related to extracellular thiol/disulfide exchange. We then examined the colloidal stability of CSand SS-DNA NPs in cell culture medium. Sedimentation assay revealed that SS-DNA NP but not CS-DNA NP could be sedimented after incubation in culture medium with cell contact. Ellman's assay confirmed the presence of extracellular thiols secreted by the cell which could reduce the disulfide bonds in SS-DNA NPs. In vitro luciferase assay indicated the level of reporter gene expression correlated with the level of extracellular thiols in two cell lines transfected with SS-DNA NP. The release of the PEG moiety from SS-DNA NP in cell culture upon disulfide bond reduction was detected by SDS-PAGE and PEG specific iodine staining. Electron microscopy and fluorescence quenching assay further suggested that dePEGylation of SS-DNA NP by extracellular thiols caused aggregation and condensation of the DNA NPs. Higher in vitro gene transfer by SS-DNA NP could be partially attributed to this aggregation, since a partial increase of reporter gene expression was observed when cells were transfected with non-PEGylated DNA NPs which also aggregated in culture medium. However, this observation was cell line dependent.

In conclusion, we report an extracellular mechanism for improving in vitro gene transfer mediated by reducible DNA NPs. Ongoing studies are focused on intracellular mechanisms. To separate extracellular mechanisms from intracellular ones, modified SS-DNA NPs are being developed with enhanced stability in extracellular environment. This is achieved by introducing steric hindrance at the carbons adjacent to the disulfide bond. The detailed mechanisms of gene transfer enhancement described by this study may guide rational design of a more efficient delivery system, which may ultimately lead to a useful gene therapy for CF and other diseases.

Supported by NIH grant 5R01DK058318. The prevalence of biofilms in human disease and in the lungs of cystic fibrosis (CF) patients constitutes a major cause of morbidity and mortality. The increased antibiotic tolerance of bacteria in biofilms is a significant problem as traditional antibiotic strategies become less effective against biofilm eradication. Pseudomonas aeruginosa infections in the lungs of CF patients are characterised by overproduction of alginate or extracellular polysaccharide (EPS) produced by the bacteria themselves. This study investigated a novel treatment strategy to determine if the susceptibility of bacterial biofilms to antibiotic/antimicrobial therapy could be modulated by treatment with defined alginate oligomers. Alginates are linear polymers of β-D-mannuronic acid (M) and/or α-L-guluronic acid (G). The ability of alginates to potentiate the activity of antimicrobial/antibiotic therapy against Pseudomonas sp. was investigated in standard MICs and, against biofilms, using an established minimum biofilm eradication concentration (MBEC) assay. Pseudomonas biofilms were pre-treated with alginate oligomers with a contiguous sequence of G residues (ca. 2600 M W ) having at least 90-95% of the monomer residues as G residues. These studies demonstrated MBEC values of a range of antibiotics including amikacin, tobramycin, oxytetracycline and ciprofloxacin were reduced (by up to three orders of magnitude) when Pseudomonas sp. biofilms were pre-treated with a 6% alginate oligomer solution. The effect of the alginate oligomers on bacterial biofilm populations was studied using live/dead staining, scanning electron microscopy (SEM) and confocal laser scanning microscopy. These studies demonstrated the ability of the alginate oligomers to effectively disrupt in vitro biofilms with loss of EPS demonstrated by SEM at higher alginate oligomer concentrations. This study highlights the potential of alginate oligomers, in combination therapy, to reduce the tolerance of established biofilms to conventional antibiotic/antimicrobial therapy. Cystic fibrosis (CF) is caused by mutation in the cftr gene, encoding for cystic fibrosis transmembrane conductance regulator (CFTR), a cAMPdependent anionic channel. The deletion of Phe508 (delF508) in CFTR first nucleotide binding domain is the most frequent mutation in CF patients in the Caucasian population. DelF508CFTR is a misfolded, rapidly degraded protein. Its rescue is a major goal for CF therapy. A recent study shows that treating Estradiol (17β E) a delF508CFTR airway cell line with 17β restores delF508CFTR functional expression, in a NHERF1 (Na-H Exchanger Regulatory Factor 1)-dependent fashion (Fanelli et al. Biol Cell 2008; 100:399-412) . In the present study we hypothized that resveratrol (a phytoalexin with a phytoestrogenic properties) may have a similar effect. The aim of this study was to test if the treatment of cells with resveratrol leads to the functional expression of delF508CFTR. The effects of resveratrol were compared to those of 17β E regarding the expression and function of CFTR/delF508CFTR. Experiments were performed on a human breast cancer cell line (MCF7, endogenously expressing wild-type CFTR and highly sensitive to estrogenic compounds) and a human pancreatic cell line (CFPAC1, endogenously expressing delF508CFTR). Immunoblot and immunocytochemistry were used to evaluate the expression of NHERF1 and CFTR/delF508CFTR. The function of CFTR was evaluated by a fluorescent dye 6-methoxy-N-3-sulfo-propylquinolinium, SPQ. Cells were treated with 17β E (50 µM) and of resveratrol (50 µM) for 2 or 18h. cAMP-dependent anionic efflux was measured in CFPAC non-confluent cells by measuring the rate of fluorescent change after perfusion of cells with a solution containing a PKA-activated cocktail.

In both cell lines (MCF7 and CFPAC) CFTR expression increased after treatment of cells with 17β E or with resveratrol. In MCF7 cells, NHERF1 expression was strongly increased by 17β E but not by resveratrol treatment. In MCF7 and in CFPAC cells, both 17β E and resveratrol increased in a time-dependent manner CFTR/delF508CFTR expression. In CFPAC cells, immunocytochemistry analysis confirmed immunoblot experiments, and strongly suggested that the resveratrol-induced increase in delF508CFTR cytosolic level is followed by partial delivery of the protein to the plasma membrane. The results from SPQ assay performed on CFPAC cells showed a rapid change in SPQ fluorescence in cells treated with resveratrol after addition of PKA-activated cocktail but not in untreated cells suggesting that delF508CFTR is functional in cells treated with resveratrol.

These results suggest that resveratrol may rescue delF508CFTR, and may be considered as a potential therapeutic agent in CF. The mechanism of action of this compound, that might be independent of NHERF1, and thus independent of an estrogenic-like effect, is under investigation.

This study is supported by Vaincre la Mucoviscidose. Background: Ciprofloxacin dry powder for inhalation (BAY q 3939) employs PulmoSphere ® formulation technology, which is designed to enhance the efficiency and reproducibility of drug delivery to the lungs. Ciprofloxacin PulmoSphere ® inhalation powder is under investigation for the treatment of chronic Pseudomonas aeruginosa airway colonization/infection in adult and paediatric patients with cystic fibrosis (CF).

Methods: This was a single centre, phase I, open-label study to assess safety, tolerability, and ciprofloxacin pharmacokinetics following BAY q 3939 administration. Adolescent patients with CF (aged 11-17 years), stable pulmonary function (FEV 1 ≥50%), BMI 15-30 kg/m 2 and confirmed P. aeruginosa airway colonization within the past 12 months received a single 50 mg dose of ciprofloxacin PulmoSphere ® inhalation powder (equivalent to 32.5 mg ciprofloxacin betaine), administered via a T-326 inhaler (Novartis Pharmaceuticals). Ciprofloxacin pharmacokinetics in plasma, urine and induced sputum were determined using non-compartmental methods.

Results: Nine patients (5 male, 4 female; mean age 14.8 years) completed the study with no severe, serious, or significant adverse events. All patients experienced mild (n=8) or moderate (n=1) dysgeusia, and one patient experienced orthostatic intolerance. No clinically relevant changes in laboratory parameters, lung function, or other vital signs were observed. Geometric mean (CV%) ciprofloxacin AUC values for sputum and serum, respectively were 204.9 mg.h/L (102.8%) and 0.71 (39.2%) mg.h/L. Geometric mean (CV%) C max values in sputum and serum were 91.1 mg/L (133.9%) and 0.14 mg/L (40.5%) and ciprofloxacin half-lives were 3.0 h (68.9%) and 4.5 h (20.4%), respectively. A mean of 45% (range 25-61%) of the inhaled dose of ciprofloxacin was recovered in urine.

Conclusions: A single 32.5 mg dose of ciprofloxacin betaine as ciprofloxacin PulmoSphere ® inhalation powder was safe and well tolerated in adolescent patients with CF. Ciprofloxacin was absorbed rapidly and exposure was >100 times higher in sputum than in serum. Systemic exposure was minimal relative to that obtained with clinically relevant doses of oral ciprofloxacin. These findings suggest microbiologically relevant ciprofloxacin exposure in the lung with minimal systemic exposure.

Supported by: Bayer HealthCare AG. Lancovutide activates a Ca 2+ -dependent alternative Cl --channel with the potential to bypass dysfunctional CFTR. This novel therapeutic approach requires a thorough dose finding study to optimize the risk/benefit ratio. Hence, a multi-centre, parallel group, placebo-controlled, double-blind efficacy and safety evaluation of three different dosage schedules of aerosolized Lancovutide was conducted. Key inclusion criteria included an age of ≥12 years and an FEV1 between 50% and 100% predicted, key exclusion criteria included ABPA, B. cepacia infection and severe liver disease. Subjects enrolled into the study were randomized to one of four treatment arms (1:1:1:1 ratio) consisting of Lancovutide (arm 1: 2.5 mg/d daily; arm 2: 2.5 mg/d every other day; arm 3: 2.5 mg/d twice a week) or placebo provided as single inhalations once daily for a duration of 8 weeks. A total of 160 subjects with mild to moderate cystic fibrosis were to be enrolled into the study to demonstrate a significant difference between the changes in the percentage of the predicted FEV1. The study consisted of a 2 weeks screening period (visit 1) followed by an 8 weeks double-blind comparative treatment period (visits 2-7) with assessment of spirometry, pulse oximetry, adverse events and quality of life using the revised CF questionnaire (CFQ-R) every 2 weeks. Thereafter, subjects were observed for additional 4 weeks without treatment (follow-up period and visit 8). Enrolment has been successfully completed in April 2009. Data cleaning is currently ongoing with subsequent closure of the data base, unblinding of the study data and assessment of the primary and secondary outcome parameters. Therefore, the study results will be available for presentation at the 23 rd NACF conference.

Supported by AOP Orphan Pharmaceuticals AG, Vienna, Austria. Acknowledgement to the study group: Bede O., Szeged; Braggion C., Florence; Colombo C., Milan; Cracowski C., Grenoble; Durieu I., Lyon; Ellemunter H., Innsbruck; Frischer T., Vienna; Gartner S., Barcelona; Hjelte L., Stockholm; Kus J., Warsaw; Lindblad A., Gothenburg; Mared L., Lund; Meyer P., Lund; Minicucci L., Genova; Pardo F., Palermo; Riethmüller J., Tübingen; Ros M., Treviso; Schuster V., Leipzig; Solyom E., Miskolc; Ujhelyi R., Budapest; Wagner T., Frankfurt. Background: In response to infection, neutrophils accumulate in the cystic fibrosis (CF) airway, yet at the tissue level the inflammatory response in the airway wall is lymphocytic. T-cells accumulate at the distal level in areas of tissue damage suggesting an involvement in respiratory or immune cell activation leading to tissue remodelling. For example, IL-17 is derived from a Th17 T-cell subset and is a potent inducer of neutrophil recruitment to the lungs. Expression of the CC chemokine and T-cell chemoattractant RANTES by the bronchial epithelium is undetectable in CF. However, platelets, which are not reported to express CFTR, are activated in patients with CF and are a source of RANTES. We previously reported that copper induces oligomerisation of RANTES, generating the minimal tetrameric structure required for in vivo chemoattractant activity.

We hypothesised that activated platelets release RANTES and induce Tcell migration across lung microvascular endothelial cells and that tobramycin and other copper chelators inhibit this response.

Methods: Human lung microvascular endothelial cells (HLMVECs) were grown on uncoated Transwell culture inserts for 2 weeks to confluency. T-cells were isolated from normal whole blood, and activated with PHA and IL-2. Platelets isolated from the same donors were activated with 1 U/ml thrombin and added to the apical or basal wells 30 min before the assay. Monolayers were treated with copper chelators, tobramycin and penicillamine, for 24 hours before adding activated T-cells to apical wells. Anti-RANTES (10 µg/ml) was added at the start of the assay. Transwells were incubated for 5 hours and migrated cells were counted using a haemocytometer.

Results: Activated platelets in the lower wells induced a significant (p < 0.01) increase in T-cell migration, which was reduced significantly (p < 0.01) by the anti-RANTES antibody. However, platelets added apically significantly (p < 0.05) inhibited migration.

Tobramycin (0.01-0.5 mM) significantly inhibited T-cell migration from 53.3±19% to 25.3±4.3 % at 0.1 mM and 15.8±6.5 % at 0.5 mM (p < 0.01). Similarly, treatment with D-penicillamine (0.01-0.5 mM) induced a significant dose-dependent decrease in T-cell migration.

There was no significant increase in lactate dehydrogenase release with tobramycin or penicillamine, and inhibitory effects were not due to cytotoxicity.

Discussion: Platelets migrate into tissues during inflammation and promote leukocyte trafficking but as yet their role has not been fully characterised. Our results demonstrate that platelet-derived RANTES is a T-cell chemoattractant in a model of the lung microvascular endothelium. It is most likely that RANTES released from activated platelets added basally binds to the endothelial monolayer forming a chemotactic gradient. Conversely, platelets added apically induced T-cell arrest, indicating that early migration of platelets across the endothelium and into tissues is essential for subsequent lymphocyte recruitment. The inhibitory effects of copper chelators on T-cell transendothelial migration indicates a proinflammatory role for copper and the potential for copper chelators to be useful anti-inflammatory agents.

Conclusion: Tobramycin has both direct and indirect anti-inflammatory properties. Cystic fibrosis (CF) is a genetic disorder characterized by mutations on the CFTR (Cystic Fibrosis Transmembrane conductance Regulator) protein resulting in a failure of CFTR-dependent Clconductance in epithelial cells. In contrast, Ca 2+ -stimulated Clsecretion is intact in CF epithelia. One possible target for drug discovery aiming at treating CF is to correct the ionic imbalance through stimulation of an alternative ionic pathway that may compensate for the deficient CFTR dependent Clsecretion. In a previous study, we identified guanabenz as an in vitro activator of Ca 2+ -dependent Clchannel (CaCC) via a Ca 2+ influx in human nasal CF airway epithelial cell line (1) . In this new study, we determined the in vitro, ex vivo and in vivo effects of guanabenz on several CF or knock-out models.

In the first part, the effect in vitro of guanabenz on CaCC was explored using single-cell fluorescence imaging in a model of human CF tracheal gland serous cell line (CF-KM4). Guanabenz induced a depolarization inhibited by DIDS, niflumic acid and NPPB. These results confirmed guanabenz as an activator of CaCC in CF cell line. Secondly, by recording shortcircuit current (Isc) measurements in mice colonic epithelium, we demonstrated that 100 µM of guanabenz induced a non-CFTR dependent increase of Isc: ∆Isc = 2.27 ± 0.09 µA/cm 2 (N = 3) in wild type mice (cftr +/+ ) and ∆Isc = 4.47 ± 0.92 µA/cm 2 (N = 3) in knock-out mice (cftr -/-). This elevation of Isc is dose-dependent giving a half maximal effective concentration EC50 of 192.2 ± 4.2 µM. Third, by evaluating the CaCC-dependent salivary secretion, we showed that guanabenz significantly potentiated the acetylcholineinduced secretion with an average of 0.20 ± 0.07 mg.g -1 (N = 7) for 100 µM of guanabenz and 0.27 ± 0.06 mg.g -1 (N = 14) for 500 µM of guanabenz in cftr -/mice.

In conclusion, we demonstrated that guanabenz is suitable for in vitro, ex vivo and in vivo activation of CaCC. Recent studies have identified TMEM16A as a glandular epithelial CaCC and suggested its role in salivary secretion (2) . We will test the hypothesis that activation of TMEM16A could be implicated in the CaCC-mediated Clsecretion induced by guanabenz.

1 Patients with cystic fibrosis (CF) and other respiratory diseases are prone to chronic inflammatory responses leading to progressive lung injury, loss of lung function, frequent exacerbations and death. In healthy people, elastase secreted by neutrophils in response to infection is neutralized by alpha-1 antitrypsin (AAT), a plasma glycoprotein present also in pulmonary tissue. In CF patients, unregulated inflammatory processes overwhelm the normal neutrophil elastase (NE) / AAT balance, leading to accumulation of NE in the lung and ultimately to tissue damage. The rationale for inhaled AAT therapy is to treat the imbalance and prevent this destructive cycle. Recently, a high purity, liquid, ready-to-use AAT was developed (Kamada Ltd., Israel) for inhalation via a customized investigational eFlow ® electronic nebulizer (PARI Pharma, Germany). The aim of this phase II study was to assess the safety and efficacy of inhaled AAT in CF patients.

Study Design: Double-Blind, Randomized, Placebo-Controlled, Repeated Dose.

Study Method: Twenty-one CF patients were randomized (2:1) to receive one inhalation per day of 80 mg active AAT or placebo. The study comprised three treatment periods, i.e., 1 day, 7 days and 28 days. Safety and efficacy variables were tested.

Results: All patients completed the study without serious adverse events (SAEs) and one patient experienced a probably related AE (mouth dryness). A decrease in sputum neutrophil percentage and in sputum NE level was detected in the AAT group, but not in the placebo group.

Conclusion: Kamada AAT inhaled once a day is safe and well tolerated in CF patients. The observed reduction of neutrophils and NE in sputum suggests an anti-inflammatory role of AAT in these patients. Further clinical trials with a larger cohort are needed to determine the effect on clinical outcomes.

The cystic fibrosis transmembrane conductance regulator (CFTR) is highly expressed in salivary tissues, and biochemical defects in saliva have been suggested previously (1) . Since salivary secretions are readily obtainable and their composition likely to be influenced by CFTR, we tested CF saliva as a source of novel disease biomarkers.

Objective: The primary goal of this study was to evaluate the salivary fluid from patients with CF and a cohort of non-CF individuals using a proteomic approach with liquid chromatography and tandem mass spectrometry (LC-MS/MS).

Methods: Based on preliminary findings in a study of 10 CF and 5 non-CF individuals, a cohort of 21 additional CF patients and 21 matched control subjects were enrolled in a prospective evaluation of the cystic fibrosis salivary proteome. Demographic data was provided at the time of saliva collection. Samples were processed and digested prior to being placed on a packed capillary tip with C18 resin. Flow rates during the separation were 350 nL/min. The samples were run on a Thermo-Finnigan LTQ-XL ion trap mass spectrometer (MS) using a full mass range of 400-2000 amu followed by the acquisition of three MS/MS spectra. All tandem MS data were analyzed using SEQUEST, X!TANDEM, and MASCOT separately and the information was combined using protein PROPHET. Final analysis was performed using Genedata Expressionist software. Identification of the 25 highest intensity peaks was performed and t-tests conducted on intensity between CF and non-CF individuals.

Results: A total of 536 peptide peaks were identified in salivary samples by LC-MS/MS. Global and robust differences in the salivary proteome among CF versus non-CF individuals were observed with 408 peptide peaks demonstrating significant differences in intensity (p-values less than 0.01) in the setting of various CFTR mutations. Confirmation of protein identity was verified by observation of multiple ion peaks from the same peptide and by Western blotting. Five proteins with the greatest degree of statistical significance included profilin (P07737), actin, cytoplasmic 1 (P60709), cystatinb (P04080), mucin-5B (Q9HC84), and keratin, type II cytoskeletal 1 (P04264). All of these had p-values of ≤0.001 when compared between the two cohorts.

Conclusion: Saliva is a readily available body fluid that contains numerous proteins that appear to discriminate CF from non-CF individuals. Further study is warranted to elucidate the identity, function, and possible role of these as biomarkers. For example, in the setting of CFTR potentiator or corrector therapy, such protein markers might ultimately furnish a useful tool for assessing efficacy. The experiments also provide a means by which the mechanisms underlying exocrine dysfunction in CF might be better understood in the future. The development of gene therapy as a treatment for cystic fibrosis (CF) lung disease has been limited in part by the lack of an animal model that presents a similar disease progression as humans. The generation of a porcine model of CF by somatic cell gene targeting and cloning to generate a CFTR null allele may provide the research community with an animal model that more closely mimics the human phenotype. We are optimizing a lentiviral vector for disease correction in this model by screening transduction efficiencies in well-differentiated primary cultures of pig airway epithelia (PAE) and wild-type pigs in vivo. Species-specific retroviral restriction factors may present barriers against lentiviral gene transfer. We contrasted the restrictive properties of human and pigs cells against both feline immunodeficiency virus (FIV)-and HIV-based vectors. Surprisingly, HIV-1 was restricted more by porcine cell lines and primary epithelia than was FIV. Using the FIV-based vector, we previously reported that the envelope glycoproteins from baculovirus (GP64), retrovirus (JSRV), coronavirus (SARS), and filovirus (Ebola) confer apical entry into differentiated primary cultures of human airway epithelia whereas VSV-G confers basolateral entry. Screening of the apical and basolateral transduction profiles of pseudotyped FIV of well-differentiated PAE resulted in gene transfer efficiency and polarity similar to that observed for human airway epithelia. In addition, we are testing the efficacy of multiple influenza A pseudotypes. Finally, we generated FIV vectors that express a variety of reporter genes including eGFP, mCherry, firefly luciferase, gaussia (secreted) luciferase, and nuclear targeted beta-galactosidase. Each of these reporters have attributes that make them suitable for measuring transduction efficiency and persistence in pig epithelial cells in vitro or in vivo. To date, our findings indicate that pseudotyped FIV lentiviral vectors confer similar tropisms in porcine epithelia as observed in human primary cultures and have implications for the selection of vectors and envelope pseudotypes appropriate for preclinical gene therapy studies in the porcine model. Background: Patients with cystic fibrosis (CF) frequently suffer from infective exacerbations caused mainly by Pseudomonas aeruginosa and in a few cases by Burkholderia cepacia. Colonisation with both of these organisms is associated with increased morbidity and mortality. However, colonisation with Burkholderia cepacia (BC) is associated with an even poorer prognosis in many patients. There are many reports in the literature on the use of inhaled antibiotics leading to increased respiratory function and decreased numbers of infective exacerbations in patients colonised with Pseudomonas. However there is limited data available on patients infected with B. cepacia.

For a number of years at the Royal Brompton Hospital nebulised antibiotics have been used as maintenance therapy for patients with Burkholderia cepacia. We are now reporting results on 15 patients treated with nebulised Ceftazidime on whom we have data for at least 1 year before and after initiation of maintenance nebulised Ceftazidime. At time of commencing treatment the organism was sensitive to Ceftazidime.

Method: A retrospective analysis was carried out in all patients colonised with BC. These patients were identified through our database. The change in lung function was calculated for the year before and after treatment with ceftazidime was given. Efficacy was analysed by assessing patients' lung function.

Of the 15 patients analysed, 7 were female and 8 were male. The average age of colonisation with B. cepacia was 22 years with a range of 18-35 years.

Average number of years from cepacia colonisation to commencement of treatment was 3.8 years with a range of between 0 -10 years.

The patients were all treated with nebulised ceftazidime at a dose of 1g bd.

Results: Of the 15 patients, 11 showed an improvement in their FEV1 following treatment with nebulised Ceftazidime. Eleven patients also showed an improvement in their FVC following treatment.

A student t test was used to analyse the results. Pre-treatment, there was a mean decline in predicted FEV1 of 13% per year, whilst after treatment there was an increase in predicted FEV1 of 7% per year (p 0.002).

Pre-treatment, there was a mean decline in predicted FVC of 9% per year, whilst after treatment there was an increase in predicted FVC of 10% per year (p 0.001).

The work has shown a statistically significant difference in rate of decline in FEV1 and FVC following treatment with 1g bd nebulised Ceftazidime, and suggests a place for inhaled ceftazidime in the maintenance therapy for those affected by B. cepacia.

Quittner, A.L. 1 ; Retsch-Bogart, G.Z. 2 ; McCoy, K.S. 3 ; Oermann, C.M. 4 ; Gibson, R. 5 ; Lewis, S. 6 ; Montgomery, B. 6 1. Univ. of Miami, Coral Gables, FL, USA; 2. Univ. of North Carolina, Chapel Hill, NC, USA; 3. Ohio State Univ., Columbus, OH, USA; 4. Baylor College of Medicine, Houston, TX, USA; 5. Univ. of Washington, Seattle, WA, USA; 6 . Gilead Sciences, Seattle, WA, USA Introduction: The Cystic Fibrosis Questionnaire-Revised Respiratory Symptoms Score (CFQ-R RSS) is a disease-specific, validated, patientreported measure of respiratory symptoms, comprising six questions. In a Phase 3, double-blind, randomized, placebo-controlled study (AIR-CF1), 164 patients with CF (age ≥ 6 years with pulmonary Pseudomonas aeruginosa and FEV 1 ≥ 25% and ≤ 75% predicted) from 67 study centers (US, Canada, Australia, New Zealand) received AZLI 75 mg three times daily for 28 days. The CFQ-R was completed at each study visit prior to any other study procedures. The primary endpoint was change from baseline at Day 28 in CFQ-R RSS. Patients treated with AZLI had significant improvement in CFQ-R RSS compared to those treated with placebo (∆9.71, p=0.0005). One of the concerns of a composite score is that deleterious effects reported for a single question may be masked by beneficial effects reported in others. We investigated whether a particular question in the CFQ-R RSS contributed to the overall positive AZLI treatment effect.

Methods: In AIR-CF1, treatment effects were examined for each CFQ-R RSS question using an ANCOVA model, adjusting for baseline disease severity (defined by FEV 1 > or ≤ 50% predicted) and baseline CFQ-R RSS item score. Child, teen, and adult questionnaires were combined for analyses. Patients < 14 years did not complete the 4th and 6th questions included in the CFQ-R RSS.

Results: For each of the six questions composing the Respiratory Symptoms Score, there was greater improvement at Day 28 for patients who received AZLI compared to patients who received placebo. Mean changes from baseline were all positive, indicating that treatment was associated with a significant improvement for each of the six symptoms; the p-values for the difference between AZLI and placebo for these six symptoms ranged from 0.001 to 0.045.

Conclusions: No single component of the CFQ-R Respiratory Symptom Score is responsible for the overall benefit of AZLI treatment in improving respiratory symptoms among patients with CF. Improvement was noted in all respiratory symptoms included in the questionnaire.

Supported by Gilead Sciences, Inc. A critical bottleneck in the drug discovery pathway for compounds that improve the biochemical synthesis and channel function of ∆F508-CFTR has been the electrophysiological evaluation of compounds. Until now only two electrophysiological methods have been available, Ussing chamber short circuit current (I SC ) measurements and the patch clamp technique.

Both of these assays are low throughput, time consuming, resource intensive and tedious. Here we describe the development of a higher throughput, simple electrophysiological functional assay that can be used to evaluate and discover ∆F508-CFTR modulators.

A change in ion channel activity is expected to alter the membrane conductance (Gm) of the cell to the permeant ion. Thus an increase in CFTR activity is expected to increase Gm for chloride. Indeed changes in Gm have frequently been used to study CFTR activity in such large cells as Xenopus oocytes. However, the small size of mammalian cells and the technical difficulties of impaling these cells with microelectrodes precludes the use of this approach for drug discovery in epithelial cells. Fortunately FRT cells express CFTR in both the apical and basolateral membranes and therefore activation of CFTR is expected to increase the transepithelial conductance (Gt) for anions. The assay we developed utilizes the expected change in Gt as a measure of CFTR channel activity. ∆F508-CFTR expressing FRT cells were grown on Costar HTS 24 well filter plates and Gt was measured using a WPI EVOMX device and modified electrodes that provide a precise, accurate and reproducible measure of Gt. The effects of compounds on CFTR expression levels (correctors) and gating (potentiation) were evaluated. Studies to date have demonstrated the measured changes in Gt are stimulated by cAMP agonists, increased by known correctors and potentiators in the CFFT compound panel and inhibited by Inh-172 and Gly-H-101. Low temperature treatment (27∞C) of ∆F508-CFTR expressing FRT cells caused the most pronounced increase in Gt upon agonist stimulation. None of these effects were observed in parental untransfected FRT cells indicating the CFTR dependence of the Gt responses in CFTR expressing cells. Depending on the number of electrode probes one uses it is possible to measure 1 to 24 wells simultaneously. Routinely, we now use a four probe manifold and four EVOMX devices to measure four wells at once. Performed in this manner the Gt responses to the sequential addition of two agonists and an inhibitor can be completed on 96 filters in less than two hours.

Thus the conductance assay provides a substantial improvement in throughput, as well as requires a minimal amount of training and capital investment in equipment compared to existing electrophysiological assays. The conductance assay and the use of 24 well filter plates provide an efficient means of evaluating CFTR modulators that should aid in the discovery of CF drugs. The simplicity of the conductance assay also means non-electrophysiologically oriented laboratories can now perform their own functional measurements of CFTR channel activity thereby accelerating studies designed to understand the mechanisms of action of CFTR modulators.

Davies, J.C. 1, 4 ; Gill, D. 2, 4 ; Griesenbach, U. 1, 4 ; Voase, N. 1, 4 ; Davies, G. 1, 4 ; Higgins, T. 1, 4 ; Innes, J.A. 3, 4 ; Boyd, C. 3, 4 ; Porteous, D. 3 The UK CF Gene Therapy Consortium is working toward a multi-dose gene therapy study, using the best currently-available non-viral gene delivery complex, and whose endpoint will be to detect clinical benefit rather than molecular or electrophysiological proof-of-principle. Based on extensive preclinical testing our selected product is pGM169, a CpG-free human CFTR plasmid with a CpG-free CMV enhancer and human elongation factor 1 alpha (hCEFI) promoter complexed with GL67A, a mixture of 3 lipids: GL67, DOPE and DMP-PEG5000. We are currently undertaking a single dose safety study (Pilot Study) because of a requirement to confirm safety of this "first-in-man" product; however, the study design has been tailored also to assess gene expression in vivo in CF lungs.

A single nebulised dose of 20 ml (53 mg pGM169 and 286 mg GL67A) is delivered by an Aeroeclipse II breath-actuated device; a nasal dose of 10% of the nebulised volume is administered on the same occasion using a standard nasal spray device. The latter allows assessment of gene expression without the sampling issues inherent in lower airway assessment, as well as anchoring to our previous clinical trials. Safety measures include physical examination, lung physiology (spirometry, pulse oximetry, lung clearance index), systemic and sputum inflammatory markers, renal and hepatic function and chest CT. We are also measuring anti-nuclear and anti-double stranded DNA antibodies and specific CFTR-related T cell responses. Measurements are made at intervals prior to dosing and during a 28 day follow up period. Gene expression is assessed by a) quantitative Taqman RT-PCR for transgene mRNA on nasal and bronchial brushings, b) anti-CFTR immunohistochemistry, and c) nasal and lower airway potential difference measurements. Given inter-subject variability, paired measurements on individuals will be obtained; bronchoscopies are being performed prior to dosing and either 2 days (n=12) or 14 days (n=12) post-dosing. Nasal PD is measured on serial visits.

To date, the 3 patients in an initial non-invasive (non-bronchoscopic) cohort have received a half dose of 10 ml; the Data Safety Monitoring Board has granted permission to proceed to the full dose. Available data will be presented on safety, tolerability and gene expression following the full 20 ml dose.

Funded by the UK Cystic Fibrosis Trust. We have identified the non-viral gene transfer formulation pGM169/GL67A for the next phase of the UK Cystic Fibrosis Gene Therapy consortium clinical programme. Quantification of pGM169-derived vector specific mRNA by real-time quantitative TaqMan RT-PCR is a key outcome measure for our clinical programme, additionally we also determine endogenous CFTR mRNA levels allowing us to determine percentvector mRNA/endogenous mRNA ratio (%VE). Cell mixing, transgenic animal and mutational analysis suggest that a gene transfer formulation that achieves >5 %VE might be expected to be therapeutic. Simplistically, the determination of %VE in clinical samples is straightforward -standard methods exist for bronchial and nasal cell sampling, mRNA preparation and TaqMan RT-PCR. However, when we evaluated the performance of these standard methods in bronchial brushing samples from volunteer CF subjects it became apparent that the Residual Sensitivity (RS) of the assay (that value of %VE required for detection and quantification of successful gene expression) was at nearly 5 %VE. It was thus essential to refine our method by systematic optimisation of every individual step in the assay with the intention of reducing the RS to levels well below that required for therapeutic benefit. During initial sample collection we evaluated the impact of bronchial and nasal brushing methods, airway cell retrieval strategies and time allowed from sample removal to cellular preservation. We found that pooling bronchial brushings dramatically improved assay RS by minimising the fixed losses associated with mRNA purification. Furthermore, the speed of initial sample processing was found to be a key factor, and we now require that it is completed with 15 minutes. Nevertheless, mRNA yields from bronchial brushing samples remain the key limiting step in the assay. Consequently, in the TaqMan RT-PCR stages, we evaluated the impact on RS and assay range and precision of alternative sample division strategieshigher RNA usage per replicate with low replicate numbers and vice versa along with the impact of multiplexing vector and endogenous CFTR assay within the same replicate. While each one of these steps in isolation had a modest impact on assay RS, the cumulative effect was highly significant. Using the optimised sample handling and assay conditions, RS in CF bronchial brushing samples improved dramatically to 0.09% (p<0.05) and RS in CF nasal brushings improved from 0.9% to 0.08% (p<0.05). In parallel, our strategy of co-purification of DNA from the same cell lysates as for mRNA purification has allowed us to also determine plasmid DNA delivery in clinical samples without affecting mRNA assay RS. Together, these improvements in the processing of clinical samples and detection of vector RNA and DNA will permit the best opportunity to evaluate the delivery and expression of our gene therapy formulations in CF patients. Primary cell culture provides an accurate and robust experimental means to approximate human airway cellular physiology in vivo. We standardized and characterized primary cell cultures using murine nasal septal or human sinonasal epithelia grown at an air-liquid interface. Septal epithelia demonstrate excellent differentiation with cilia in approximately 90% of cultured cells, and ciliary beat frequencies stimulated by physiologic agonists, such as ATP and forskolin. Septal epithelial cultures also have substantially greater CFTR activity in Ussing chamber analysis than comparably processed tracheal monolayers, and allow direct comparisons of otherwise genetically identical (congenic) animals without the confounding effects attributable to mixed genetic backgrounds. Nasal primary cells can provide mechanistic comparisons with the nasal potential difference assay, a mainstay of CFTR detection in vivo. The culture model may thus be ideal for testing activity of CFTR correctors and potentiators, and determining whether such agents restore normal ion transport by potentiating the CFTR activation pathway.

In the present study, tissues from wild type, heterozygous ∆F508, and homozygous ∆F508 murine nasal septae (Cftr tm1Kth mice) were cultured at an air-liquid interface on filter supports to confluence and full differentiation. CFTR expression was measured with quantitative RT-PCR. Native tissues were also harvested for evaluation. Homozygous ∆F508 cultures exhibited no detectable CFTR activity following low temperature correction. Quantitative PCR (reported as relative mRNA levels ± S.D.) showed dramatic differences in CFTR mRNA between homozygous (6.5 ± 2.5), heterozygous (68.9 ± 23.1), and wild type cultures (129.9 ± 56.3, p < 0.0002).In addition, there was a 4.5-fold (homozygous) and 2.5-fold (heterozygous) decrease in CFTR mRNA compared to wild type in nasal tissues from mice.

Prior studies concerning Cftr tm1Kth transgenic mice have indicated tissue specific differences in CFTR expression, and in particular, lower levels of ∆F508 CFTR mRNA in the upper intestine compared to wild type mice. Normal levels of CFTR mRNA were reported previously in ∆F508 murine lung, testes, and submaxillary glands. In the present experiments, CFTR expression was reduced in ∆F508 nasal epithelium, and markedly repressed in primary tissue culture. Whether these differences are mediated by higher expressivity in a pure epithelial cell population (compared to mixed cells from native tissues), decreased CFTR transcription, or message instability will require further investigation. However, our findings indicate significant limitations of airway cultures from Cftr tm1Kth mice, and those previous studies of ∆F508 CFTR from the tm1 Kth background (generated by exon replacement) should be reevaluated in this light. For example, potentiator activation profiles observed in ∆F508 CF mice may underestimate potency as a result of diminished levels of mRNA. Emerging transgenic ∆F508 CF animal models (pig, ferret) should also be evaluated for decreased gene expression in both cell culture models and native tissues. Gene therapy is being evaluated for treatment of chronic cystic fibrosis (CF) lung disease. When plasmid DNA expressing CFTR is complexed with cationic liposomes or polymers, it can be delivered to the lungs of mice and sheep via aerosol, and CFTR mRNA can be detected in lung tissue. Recently, the CpG-free plasmid pGM169, which can result in persistent, inflammation-free, expression of human CFTR via the hCEFI promoter in the mouse lung (Hyde et al. Nat Biotechnol 2008; 26:549) was selected for clinical study. Currently, the UK Cystic Fibrosis Gene Therapy Consortium is conducting a Phase I clinical trial to evaluate the safety of aerosol delivery of a pGM169/GL67A liposome formulation to the lungs of CF patients. Although the primary aim of the clinical study is to evaluate safety, we are also taking the opportunity to measure gene expression in nasal and bronchial brushings. To quantify transgene expression in clinical samples, traditional 2-step TaqMan RT-PCR is routinely used as a reliable and sensitive method. Using this type of assay we are capable of quantifying pGM169 mRNA and endogenous human CFTR mRNA down to 25 and 10 copies/µl of input RNA respectively, however below this threshold the samples are non-quantifiable and are reported as negative. To improve the opportunity to detect low levels of mRNA in clinical samples we developed a 3-step TaqMan RT-PCR assay, which includes an additional nested PCR step. This resulted in improved sensitivity by facilitating detection of low concentration pGM169 mRNA mimics that were previously undetectable by 2-step TaqMan RT-PCR. To model expression of pGM169 in the human lung, sheep lung tissue samples were collected following aerosol delivery (32mg pGM169). When vector-specific CFTR mRNA was assayed in the sheep samples, 25% of samples reported as negative by 2-step TaqMan PCR, had a strong (though non-quantifiable) vector signal when assayed by 3-step TaqMan PCR. Similarly, RNA from human Air Liquid Interface cells transfected with pGM169 resulted in detection of vector signal in 50% of previously "negative" samples. Thus application of the 3-step RT-PCR assay to RNA samples below the quantifiable threshold demonstrated that these tissue samples actually contained vector RNA signal. The ability to detect low levels of vector-specific mRNA in heterogeneous tissue samples indicates that this assay is ideal for measuring transgene expression in clinical samples, where availability may be limited. Efficient transduction of the airway epithelium and long-term expression of the transgene are essential elements for successful cystic fibrosis (CF) gene therapy. Parsons et al (2008) showed that action of the natural detergent lysophosphatidyl choline (LPC) to open tight junctions between cells allowed improved access to viral receptors on the basolateral surface of the airway epithelium in mice. We have previously shown that bronchoscope-guided, targeted lobar aerosolization of a mixture of LPC and Helper Dependent Adenovirus (HDAd) into nonhuman primate lungs results in uniform, high level pulmonary transduction. Our ultimate goal is to use LPC as the carrier in delivery of HDAd for gene therapy in CF patients. In this study we investigated the effects of LPC on lung mechanics of non-human primates.

Methods: Baseline pulmonary function (PF) in 5 baboons were measured using the Flexivent respiratory mechanics platform (Scireq). Animals were intubated with a cuffed endotracheal tube, placed under general anesthesia, and their lungs were inflated to 30cm of H 2 O prior to each of the PF measurements. Following baseline measurements, LPC was aerosolized using two different concentrations and volumes which we have previously shown to result in significant pulmonary transduction by HDAd. Two ml/lobe of 0.1% LPC was delivered into each of the six major lung lobes of baboons using an intracorporeal nebulizing catheter (AeroProbe) inserted into a bronchoscope. PF measurements were taken after LPC administration and every 15 minutes thereafter for up to 2.5 hours to determine changes in lung mechanics. A final PF measurement was taken at 24 hours post LPC administration for those animals with lung functions that did not return to baseline at the end of the 2.5 hour testing time. Three weeks after administration of 0.1% LPC, 0.05% LPC was administered to the same 5 animals and PF measurements were obtained as previously described.

Results: Maximal PF changes occurred 15 minutes following LPC administration for both 0.1% and 0.05%LPC. Mean respiratory resistance (n=5) increased 307% ± 146 % (S.D.) above baseline and compliance decreased 53% ± 9.3% below baseline after the delivery of 0.1% LPC. Pulmonary function changes returned to baseline within 24 hours of LPC administration. When 0.05% LPC (n=5) was administered to the same baboons, their mean respiratory resistance increased 193% ± 55.4% over baseline and compliance decreased 66% ± 9.4% from baseline. However, respiratory resistance in these animals returned to normal in less than 3 hours. No significant changes in FIO 2 or SpO 2 were observed following administration of either 0.1% or 0.05% LPC.

Conclusion: LPC at 0.1% and 0.05% is associated with increased respiratory resistance and decreased compliance measurements when delivered by aerosol to the lung. Maximal changes in PF occur within 15 minutes of delivery. These results indicate that LPC delivered by aerosol to the airways of the lung produce significant but temporary pulmonary function changes in non-human primates in a dose-dependent manner. Bacterial biofilms contribute to the reduction in effectiveness of many antibiotics in chronic infections such as cystic fibrosis (CF) lung infections, chronic wound and sinus infections, endocarditis, and medical device infections. Panaecin™ (gallium ) is a novel anti-infective with demonstrated efficacy against many gram-negative and gram-positive bacteria growing under planktonic or biofilm conditions, such as P. aeruginosa, Burkholderia cepacia and methicillin-resistant Staphylococcus aureus (MRSA). Furthermore, Panaecin is effective in the presence of CF sputum against tobramycin-and aztreonam-resistant P. aeruginosa that chronically colonize the lungs of CF patients, which makes it a promising drug candidate (1) . Panaecin™ competes for iron in critical enzymatic pathways in bacteria such as DNA synthesis, metabolic conversion, electron transport and oxidative stress defense. We report the lung residence time of Panaecin ™ following an intratracheal instillation in rats which supports a once a day, or even less frequent, administration by inhalation. A gallium (III) formulation was prepared and instilled into the lungs or injected via IV in male Sprague-Dawley rats. At predetermined time points, blood samples were taken, the animals were euthanized, the lungs lavaged and excised. Gallium (III) was extracted from the lung tissue, bronchial-alveolar lung fluid (BAL), and serum, and subsequently quantified using flameless atomic absorption spectroscopy. An approximate 5-fold greater concentration was observed in lung tissue within 1 hour after intratracheal administration of 2 mg/kg of the inhalation formulation of gallium (III) compared to intravenous injection of the same dose. Approximately 50% and 95% was eliminated from lung compartments within 60 minutes following IT instillation or IV injection, respectively. The remainder of the dose was eliminated with a half-life of 24 hours. This compares well with the reported clinical findings following inhalation of a gallium-67 solution by healthy volunteers in which approximately two-thirds of the dose remained in the lung after 24 hr (2) . In comparison, the reported lung clearance half-life for tobramycin following IT instillation of 10 mg/kg to rats is 147 minutes (3).

Pulmonary administration of the inhalation formulation of gallium nitrate results in higher local lung concentrations than intravenous administration. Comparisons to reports from the literature suggest that the advantage of pulmonary administration over oral delivery is even greater. The development of an inhaled gallium formulation would have many benefits for the treatment of respiratory infections, and compliance could be improved by a once daily administration.

References It has long been proposed that inhaled alginate lyase enzymes capable of degrading the exopolysaccharide alginate would render mucoid, cystic fibrosis (CF) associated P. aeruginosa infections more susceptible to the human immune response and antibiotics, resulting in an overall improvement in lung function. The therapeutic candidate Sphingomonas sp. A1-III alginate lyase (A1-III) has shown significant potential for bacterial alginate degradation, but unfortunately is predisposed towards excessive immunogenicity as a result of its bacterial origin. To address this problem, we have genetically engineered an A1-III enzyme containing a single cysteine substitution permitting controlled, orthogonal, maleimide-thiol conjugation to polyethylene glycol (PEG). This rational approach has produced a PEGylated A1-III enzyme with 48% decreased antigenicity, as assessed by enzymelinked immunosorbent assays (ELISA), and a 44% decreased in vivo immunogenicity in a rabbit model. Furthermore, antibody binding studies with a naïve repertoire of human single chain variable fragments displayed on yeast showed a 90% reduction in human scFv binding for the PEGylated variant compared to the native enzyme, suggesting that the de-immunization effect translates to the human immune system. In addition to reduced immunogenicity, the PEGylated A53C variant was found to possess catalytic activity equivalent to the native enzyme when assessed with a soluble brown seaweed alginate substrate. Importantly, PEGylated A53C was found to be 80% more active than the native enzyme when assayed with alginate purified from a mucoid P. aeruginosa clinical isolate, and in vitro studies demonstrated that 0.25 mg/ml doses of the PEGylated enzyme were able to reduce established mucoid P. aeruginosa biofilms by 30% relative to a notreatment control. Finally, we have observed that addition of a C-terminal his-tag to the enzyme results in a 300% increase in its maximum reaction velocity. It is a tantalizing possibility that his-tag mediated metal-ion interactions may improve the alginate degrading capacity of therapeutic candidates. As a whole, these results demonstrate that genetically and chemically modified A1-III alginate lyase exhibits therapeutically relevant, enhanced alginate degrading activity while simultaneously possessing decreased potential for eliciting an immune response in patients. A library of silver N-heterocyclic carbene complexes has been synthesized and the complexes are under investigation as potential antimicrobial agents for aerosol delivery. Although a silver carbene complex (SCC) designated SCC1, a methylated caffeine silver(I) acetate molecule, has demonstrated efficacy in treating respiratory infections in mouse models when dosed every 12 hours, there remains a need for packaging, protection and targeted delivery of Ag(I) in order to increase dosing intervals and hence, patient compliance. We sought to develop nanoparticles loaded with SCC1 in order to provide depot delivery of the drug, but for technical reasons, SCC1 could not be easily incorporated into nanoparticles. In this study, we developed shell crosslinked knedel-like (SCK) nanoparticles, loaded with Ag(I) both as 1-hexyl-3-methyl-4,5-dichloro-imidazole-2-ylidene silver (I) acetate (SCC10) and as free Ag(I), as an advanced antimicrobial device, designed to encapsulate and protect Ag(I) from salts and other biological agents, while retaining the potential for functionalization to achieve targeting and imaging capabilities. The SCKs were constructed by the supramolecular assembly of amphiphilic block copolymers, poly (acrylic acid)-bpolystyrene (PAA-b-PS), into micelles, followed by covalent crosslinking throughout the shell layer to afford discrete nanostructures having a hydrophobic core domain and a hydrophilic shell region. The silver-bearing nanoparticles were characterized and their antimicrobial activities against strains of Pseudomonas aeruginosa isolated from cystic fibrosis (CF) patients were evaluated both in vitro and in a mouse model. The minimal inhibitory concentrations of SCC10 against P. aeruginosa ranged from 1-6 µg/mL. The activity of the silver-bearing SCK constructs against a representative strain of P. aeruginosa (strain PA M57-15; MIC [SCC10] = 1 µg/mL) was measured by monitoring optical density (650 nm) in a microplate spectrophotometer 6 h after treatment with constructs equalized for silver content as determined by ICP-MS. Independent of the silver loading method, decrements in the growth of P. aeruginosa PA M57-15 were observed at [Ag + ] of 2-4 µg/mL and growth was completely inhibited at [Ag + ] of 8 µg/mL. Intranasal inoculation of mice with P. aeruginosa (PA M57-15) followed by 2 nebulized doses of water, SCC10-Ag-SCK or SCC10-SCK constructs delivered at 1 and 25 hours after inoculation resulted in a 40 and 60% survival advantage, respectively, compared with water highlighting the importance of the sustained release of Ag(I) from the SCC10 loaded into the core domain versus the immediate release of shell-loaded Ag(I). These silver-loaded SCK nanoparticle delivery systems exhibited potent antimicrobial activities both in vitro and in vivo and may provide effective, sustained antimicrobial therapy for treatment of CF pulmonary infections. Pseudomonas aeruginosa (PA) is a frequent cause of respiratory infections in individuals with cystic fibrosis (CF). Azithromycin, which has no activity against PA at achievable serum concentrations, is associated with improved outcomes in patients who have chronic PA lung infection, although the mechanism is unclear. Unlike antipseudomonal beta-lactams and aminoglycosides, which penetrate poorly into mammalian cells, azithromycin is concentrated to high levels within phagocytes and may potentially exceed the MICs of PA strains in this compartment. We therefore postulated that during infection azithromycin eradicates intracellular PA, augmenting the efficacy of conventional antipseudomonal antibiotics. Using an in vitro infection model, PA was incubated with J774 macrophage-like cells in the presence of amikacin with or without azithromycin. Recoverable internalized bacterial CFUs were enumerated after 24 hours of incubation by plating. In vivo experiments were performed using a well-characterized nasal aspiration mouse model of acute pneumonia. Mice were treated with amikacin with or without azithromycin. In separate experiments, pulmonary phagocytes were purified from infected mice by gradient centrifugation, and viable internalized bacteria enumerated from cell lysates. In addition, disease severity and survival were measured over 96 hours. In vitro assays confirmed that a small portion of the PA inoculum persisted within J774 macrophage-like cells. The addition of azithromycin to these assays resulted in a significant reduction (p<0.05) in the number of surviving intracellu-lar bacteria after 24 hours of infection compared to amikacin or piperacillin alone. In the mouse model of pneumonia, a significant improvement in clinical well-being and survival was observed in mice treated with amikacin plus azithromycin relative to amikacin alone (p<0.01), untreated controls (p<0.001) and azithromycin alone (p<0.001). Significantly fewer viable bacteria were recovered from pulmonary phagocytes of mice treated with azithromycin plus amikacin compared to amikacin alone (p<0.03). The improvement in survival with addition of azithromycin was not observed when an isogenic PA strain that remained largely extracellular was used for infection. Thus both in vitro and in vivo models demonstrated decreased severity of infection with the addition of azithromycin to standard PA therapy, and in both cases improved outcomes were associated with decreased numbers of viable bacteria within phagocytes. These results suggest that eradication of intracellular bacteria may be another mechanism by which azithromycin contributes to improved outcomes in CF patients with PA airway infections.

Supported by the Northwestern Memorial Foundation, the American Lung Association, NIH, and Liv For a Cure. Background: New antibiotics with high activity against biofilm producing bacteria are needed to treat chronic airway infections. Fosfomycin/tobramycin for inhalation (FTI) is a novel inhaled antibiotic combination consisting of a 4:1 (wt/wt) ratio of fosfomycin and tobramycin. FTI is active against a wide spectrum of pathogens grown under planktonic conditions, but its activity against bacteria grown in biofilms has not yet been characterized. The objective of this study was to compare activities of FTI to antibiotics available or in development against P. aeruginosa biofilms grown on human CF bronchial epithelial monolayers (CFBE41o-). Additionally, we evaluated the effects of antibiotics on inflammatory markers associated with P. aeruginosa infection of CFBE41o-cells.

Methods: FTI, fosfomycin, tobramycin and aztreonam were evaluated at total concentrations ranging from 1-4096 µg/mL. Inhibition of biofilm production was determined 16 h after adding antibiotic and bacteria to CFBE41o-monolayers. P. aeruginosa persistence in biofilms was assessed by incubating established P. aeruginosa biofilms with antibiotics for 16 h. In both experiments, biofilm colony forming units (CFU) were determined after washing away planktonic bacteria. CFBE41o-cytotoxicity was determined by measuring lactate dehydrogenase (LDH). The effects of antibiotics on the release of TNFα, IL-1β, IL-6, IL-8, MIP-3α, RANTES, IP-10, GRO-α, FGF-basic, and VEGF from CFBE41o-cells were evaluated by a multiplexed assay (Affymetrix Inc., Santa Clara, CA).

Results: Inhibition of biofilm formation was achieved at 64 µg/mL FTI, 16 µg/mL tobramycin and 1024 µg/mL aztreonam. Fosfomycin did not inhibit biofilm formation at concentrations ≤1024 µg/mL, but dramatically reduced P. aeruginosa biofilm-mediated CFBE41o-cytotoxicity. Sub-bactericidal concentrations of FTI (16 µg/mL) and tobramycin (4 µg/mL) also reduced cytotoxicity of CFBE41o-cells. Release of IL-6 from infected CFBE410-cells was inhibited by 1024 µg/mL of fosfomycin. P. aeruginosa persistence in biofilms was reduced 4 log 10 by FTI and tobramycin at 256 µg/mL, and both demonstrated superior activity compared to fosfomycin and aztreonam.

Conclusions: FTI inhibited biofilm formation and reduced persistence of P. aeruginosa in established biofilms. In addition, there is a suggestion of cytoprotective and anti-inflammatory effects that warrant further evaluation.

Supported by Gilead Sciences, Inc. The risk of selecting for drug resistant pathogens is inherent to the use of all antibiotics. The development of antibiotic resistance in patients with CF is of particular concern because complete clearance of lung pathogens is difficult to achieve and PA can develop resistance to multiple antibiotics. AZLI is an anti-pseudomonal antibiotic with a lysine excipient that has significantly reduced patient sputum PA density in two 28-day course, Phase 3, double-blind, randomized, placebo-controlled studies. We investigated whether repeated treatment courses of AZLI led to increases in the MIC of multiple antibiotics to PA isolates.

Methods: In a Phase 3, 18 month, open-label, follow-on study (AIR-CF3) patients could receive up to nine courses of AZLI therapy, 28-days of AZLI (BID or TID) followed by 28 days off treatment. Patients attended up to 20 scheduled clinic visits (every 28 days). At each study visit, sputum samples were collected and processed according to the CF Foundation standards and cultured for PA. Depending on the presence or absence of multiple morphotypes, one or more PA CFUs were tested for antibiotic susceptibility against 10 antibiotics using a microbroth dilution technique.

Results: At the end of the last AZLI course of AIR-CF3, the MIC50 and MIC90 of all PA isolates tested was unchanged (±2 fold difference) from the baseline values of 4µg/mL and 128 µg/mL, respectively. The subgroup of PA isolates with the highest MIC 50 and MIC 90 to aztreonam from each patient demonstrated a similar result. The MIC 50 and MIC 90 of PA isolates collected from patients using AZLI TID increased 3 or fewer times (>4 fold change, µg/mL) from baseline throughout the 9 treatment courses. PA isolates collected from patients using AZLI BID demonstrated increases in their MIC 50 and MIC 90 more frequently than PA isolates collected from patients undergoing AZLI TID therapy. The percentage of patients with PA isolates resistant to all β-lactams did not consistently increase over baseline during the nine treatment courses. The same is true for the percentage of patients with PA isolates resistant to all aminoglycosides tested. Further, as high as an 8 fold decrease in the tobramycin MIC 90 of PA isolates with the highest MIC to tobramycin was observed during AIR-CF3.

Conclusion: PA isolates collected from patients who received up to 9 courses of AZLI over an 18 month period did not display persistent increases in the MIC of aztreonam. AZLI TID therapy appears to have fewer increases in the MIC of aztreonam in PA than AZLI BID therapy. Increasing resistance over time to antibiotic classes was not observed in PA isolates while AZLI therapy appears to decrease the resistance of PA to tobramycin.

Supported by Gilead Sciences, Inc. Chronic infections with this strain, designated CNAX, were associated with a rapid clinical decline.

Goals: The goals of this study were to: 1. determine the impact of A. xylosoxidans infections on clinical outcomes; 2. compare outcomes of patients infected with the CNAX strain to non-CNAX strains of A. xylosoxidans; and 3. update outcomes associated with the CNAX strain.

Methods: Culture specimens were obtained via sputum or BAL samples. Confirmation of A. xylosoxidans was made by 16S rRNA RFLP analysis; strain genotyping was performed by a rep-PCR based method.

Patients in our pediatric center with lung function data from 2001-2007 were included. A linear mixed effects model was used to determine if the annual rate of decline in FEV 1 % predicted (FEV 1 ) differed for patients with CNAX and non-CNAX strains. Covariates (using backwards elimination) included age, sex, genotype, diabetes status, chronic infections and insurance status. Least squares means were used to compare mean FEV 1 for CNAX and non-CNAX strains.

Results: Twenty-five patients had chronic infection with A. xylosoxidans; nine and sixteen patients were infected with the CNAX strain and non-CNAX strains, respectively. Two additional patients with chronic CNAX infections were identified in our adult CF center, but not included in the lung function analysis. Patients with CNAX infections had a lower mean FEV 1 and a faster rate of decline in FEV 1 compared to patients with infections with non-CNAX strains (Table 1 ). There were no differences in the mean FEV 1 (p = 0.158) or the yearly rate of decline in FEV 1 (p = 0.22) when patients with non-CNAX A. xylosoxidans were compared to patients without A. xylosoxidans. Other covariates associated with a faster rate of decline in lung function were chronic infections with B. cepacia and abnormal diabetes status.

Combining our pediatric and adult outcomes, ten of eleven patients in the original CNAX cohort died or received lung transplantation within five years of CNAX infection. Two additional patients had positive cultures with the CNAX strain in our adult center within the past two years but have not had subsequent positive cultures.

Conclusions: Chronic infection with the Cincinnati strain of A. xylosoxidans (CNAX) is associated with a rapid decline in clinical status when compared to infection with non-CNAX strains of A. xylosoxidans. Chronic infection with non-CNAX strains of A. xylosoxidans is not associated with adverse clinical outcomes. (MRSA) is cultured from 20% of cystic fibrosis (CF) patients. We have previously shown that persistent MRSA infection is associated with a more rapid lung function decline in children and adolescents, but not adults. In non-CF populations, MRSA has been associated with increased mortality. Despite the increasing prevalence of MRSA in CF patients, there have been no studies examining the impact of MRSA on survival. We hypothesized that MRSA, after adjustment for severity of illness, would be associated with increased mortality.

Methods: Cohort study of individuals in the US CF National Patient Registry aged 6-45 from 1996-2007. To create a cohort with newly detected MRSA we excluded anyone with: MRSA in the first two years, less than two years of observation, or those with less than two cultures in the first two years. MRSA was a time-dependent variable that could be positive or negative in any given year. Time at risk began at the age of 7 and entry time was the age when an individual entered the cohort. Transplant patients were censored at the time of tranplant. The primary outcome was time from entry until age of death from any cause. For the comparison of unadjusted survival rates, Kaplan-Meier methods were used. Time-updated Cox regression models were used to calculate adjusted hazard ratios of death because the differences between MRSA and non-MRSA patients could confound a comparison of survival patterns; confounders in this model were sex, year, insurance status, FEV 1 , CF related diabetes, pancreatic status, Pseudomonas, and Burkholderia cepacia Complex.

Results: During the study period there were: 19,445 CF individuals who met the inclusion and exclusion criteria, 2,416 deaths, and 5,000 individuals that cultured MRSA. The average follow-up time was 6.5 years and median age of death was 35.5 years. Patients with MRSA had significantly shorter survival (p<0.001). The unadjusted hazard ratio of death associated with MRSA detection was 1.44 (95% CI 1.26 to 1.64). After adjustment for severity of illness, the hazard of death was still significantly increased (hr 1.27; 95% CI 1.09 to 1.47; p=0.002). To further address the question if MRSA contributes to mortality or is just a marker of severity of illness, we investigated the effect of modeling MRSA as a time-lagged variable. The hazard ratio of death was 1.38 (95% CI 1.17 to 1.63; p<0.001), adjusted for confounders.

Conclusions: This 10-year analysis of 19,445 CF individuals from the national patient registry suggests MRSA is associated with increased mor-tality, even after adjusting for severity of disease. Furthermore, the risk of death is increased by MRSA detection one or two years before death, implying MRSA is not just cultured in sick individuals in the last year of life. Increased mortality was not only seen in children and adolescents, but adults as well. Therapeutic trials should be undertaken to ascertain whether the morbidity and mortality associated with MRSA can be reversed.

Supported by a Clinical Research Fellowship Award from the CF Foundation. Objective: Burkholderia cepacia complex (Bcc) bacteria cause devastating infections in individuals with cystic fibrosis (CF). A DNA microarray has been designed to the genome of B. cenocepacia strain J2315, and has provided an ideal opportunity to explore the global transcriptomic response to changes in environmental conditions likely to occur when Bcc bacteria from the environment come into contact with the CF lung. Using this microarray we aimed to identify the genes expressed and adaptations made that permit Bcc bacteria to infect and persist in the CF lung.

Methods: A set of microarray experiments was performed investigating the response of B. cenocepacia J2315 to conditions relevant to the CF lung environment. The parameters investigated were: a temperature shift form 20 o C to 37 o C, a decrease in oxygen concentration from 20% to 6%, a decrease in iron concentration from 2.4 to 0.4 ppm, and a pH shift from 7.0 to 5.5. Microarray experiments for comparison of stationary phase with log phase cultures were also performed to clarify the transcriptomic response seen at different B. cenocepacia growth phases.

Results: The largest overall changes in gene expression were found to occur in response to a decrease in oxygen concentration and entry into stationary phase; there was also significant overlap between genes upregulated under these conditions. Many overlapping genes were also upregulated (to a lower extent) during the temperature shift to 37 o C, indicating a potential role in the general stress response. Several of the upregulated genes had previously been implicated in virulence and included: the zinc metalloprotease, zmpA, fimbrial proteins and most conspicuously aidA, a protein of unknown function necessary for B. cenocepacia virulence in the Caenorhabditis elegans infection model. A non-coding RNA with homologies to the 6S RNA of Escherichia coli was also prominent as an intergenic region upregulated under all stress conditions. The main response to reduced oxygen concentration was a strong upregulation (up to 200-fold) of a large cluster comprising 47 genes (encoding stress proteins and a cytochrome c) that were adjacent to a genomic island. Several other paralogous cytochrome encoding genes underwent strong expression changes in response to the alteration in oxygen concentration. Increased temperature caused B. cenocepacia to upregulate chaperonins, heat-shock proteins, and also transposases. At reduced iron concentration virtually all genes and gene clusters known to be involved in iron transport were upregulated.

Conclusions: Transcriptomic analysis of B. cenocepacia J2315 showed that genes constituting a general stress response in B. cenocepacia are upregulated in most conditions relating to CF lung environment. The microarray reference data set has enabled the identification of the most important B. cenocepacia genes regulated in response to the environmental parameters which alter during infection. The large dataset can be used to inspire future investigative approaches aimed at preventing and treating Bcc infection in CF. This research was supported by grant MAHENT06V0 from Cystic Fibrosis Foundation Therapeutics Inc. Quorum sensing (QS) is a bacterial signalling mechanism that allows communication between bacteria and the coordination of group behaviour. In gram-negative bacteria, the QS signals are acyl-homoserine lactones (AHLs) that bind to transcriptional regulators and modulate gene expression when present in sufficient quantities. QS systems in the cystic fibrosis (CF) pathogens, Pseudomonas aeruginosa and Burkholderia cepacia complex (Bcc), regulate expression of many virulence factor genes thought to contribute to acute and chronic infections. Despite convincing evidence for a role for QS in virulence, there have been several reports of lasR QS mutants of P. aeruginosa being recovered from airways of chronically infected CF patients. P. aeruginosa lasR mutants have been demonstrated to have a growth advantage over wild type strains in some conditions suggesting that QS regulated virulence factors may be more important in acute than chronic infections. The objectives of this study were to determine if similar adaptations in the CepRI QS system occurred over time in Bcc chronic infections. Genetically-related sequential isolates from 12 patients chronically infected with B. cenocepacia and 10 patients infected with B. multivorans were analyzed for AHL production. The mean collection time between the paired sequential isolates was 6.2 years for the B. cenocepacia strains and 5.7 years for the B. multivorans strains. All isolates with the exception of one B. cenocepacia strain produced AHLs suggesting that the cepRI genes were functional. The one AHL negative isolate was found to contain an insertion sequence in the cepR gene. Nine of the remaining 11 B. cenocepacia paired isolates had the same or higher AHL production in the later isolate than the original isolates. All but one of the B. multivorans later isolates produced more AHL than the original isolate from the same patient. Competitive growth assays were performed with the B. cenocepacia wild type (WT) and cepR mutant clinical isolates in an agar bead model of chronic respiratory infection and in various culture media. The growth competitiveness between the WT and cepR mutant isolate varied depending on the culture medium; however, 99% of the colonies recovered from lungs 10 days post infection in the animal model were WT, whereas the initial inoculum contained 52.5% WT. These data suggest that there is a selective advantage for AHL production and QS mediated gene regulation in Bcc Pseudomonas aeruginosa chronically colonizes the lungs of individuals with cystic fibrosis (CF). Evidence suggests that this infection process involves the transition of P. aeruginosa from an acute, virulent phenotype to a more chronic lifestyle typified by the formation of biofilms within the airways. We have previously developed a tissue culture model system for the development of P. aeruginosa biofilms directly on cultured human CFderived airway epithelial cells. Using this model, we found that the magnesium transporter MgtE influences the cytotoxicity of the bacteria toward the epithelial cells. Mutation of the mgtE gene led to increased cytotoxicity, and overexpression of this gene resulted in decreased toxicity. Using in vitro assays, we discovered that the mgtE mutant strain secreted a greater amount of toxins through the type III secretion system (T3SS). Mutational analysis and transcriptional studies of T3SS structural, effector, and regulatory genes suggested that the MgtE protein interacts with the regulatory network that governs T3SS gene transcription, likely by controlling the activity of the master regulatory protein ExsA. Intriguingly, although MgtE can transport magnesium, the effects of MgtE on cytotoxicity appeared to be magnesium-independent. These studies demonstrate that by varying the levels of MgtE in the bacterial membrane, P. aeruginosa can modulate its toxicity towards epithelial cells. This control of cytotoxicity may influence the acute to chronic lifestyle transition enacted by P. aeruginosa during infection of the CF lung. Thus, these studies suggest that future therapies might target MgtE in order to disrupt P. aeruginosa biofilms and influence bacterial toxicity. Here we performed an unbiased signature tagged mutagenesis (STM) study and for the first time we exploited a positive selection screening in a mouse model of chronic infection. As read-outs of virulence in the agar beads lung infection model, we tested P. aeruginosa persistence versus clearance to select for mutants with increased virulence. Ninety-six pools of 72 mutants each carrying a different signature tag were exposed to murine airways and viable P. aeruginosa were recovered after 14 days. After three rounds of screenings with mixed mutants, followed by a round of screening with single mutants, 16 STM mutants were selected as able to establish chronic infection in higher percentage of mice when compared to PAO1293 strain (STM mutants vs PAO1293: 80-100% vs 12.5%; Chi square analysis: PAO1293 vs STM mutants p<0.05).

Sequence analysis identified insertion into 16 distinct ORFs belonging to hypothetical, unknown, unclassified proteins (3), motility and attachment (4), putative enzymes (1), transport of small molecules (3), amino acid biosynthesis and metabolism (1), energy metabolism (1), secreted factors (1), chaperones and heat shock proteins (1) . In addition, one insertion was found in an intergenic region. Phenotypic characterization included twitching and swarming motility, mucoidy, LasR phenotype, hemolysis and autolysis, proteases secretion, biofilm formation and pyocyanine production. Five STM mutants were defective for twitching and/or swarming motility, two mutants have a pyocyanine defect and four mutants have differences in biofilm formation when compared with PAO1293 strain. None of the P. aeruginosa mutant strains have a mucoid phenotype. Next, to test whether phenotypic differences affect the ability of mutants to invade human respiratory epithelial cells, we infected A549 cells. The majority (11) of mutants were found to be statistically more invasive than the referent PAO193 strain.

Sequence alignment of the 16 novel functions identified in PAO1 against six sequenced P. aeruginosa genomes including those of clinical origin (PA2192, C3719, PAC2, PA14, PA7, LESB58), revealed major sequence variability in genes involved in quorum sensing, motility and attachment, amino acid biosynthesis and metabolism, chaperones and heat shock proteins and unknown genes.

Overall, we identified novel genes which play a critical role in the pathogenesis of P. aeruginosa chronic infections and may serve as targets for novel therapies.

Supported Pseudomonas aeruginosa is a gram-negative opportunistic pathogen that chronically infects the lungs of a majority of individuals with cystic fibrosis (CF). In the lung, P. aeruginosa produces a number of virulence factors including hemolytic phospholipase C (PlcH), which breaks down phosphatidylcholine (PC), a major component of lung surfactant. In addition to the direct toxic effects of PlcH on the host, we have shown that PC degradation also enhances P. aeruginosa biofilm formation, which may contribute to decreased host clearance and increased antibiotic resistance. Gene expression analysis of P. aeruginosa grown in the presence of PC-rich bovine lung surfactant has allowed us to identify transcripts indicative of PlcH production and PC degradation. Bacterial RNA isolated from the sputum of CF patients was analyzed by quantitative RT-PCR. These studies provided evidence that PC-degradation likely occurs in the lung during chronic infections, and preliminary data show that plcH mRNA levels may correlate with decreased forced expiratory volume (FEV 1 ). We have discovered that miltefosine, a commercially available alkylphosphocholine currently in therapeutic use, decreases PlcH activity towards its substrates, and we have found that addition of 50µM miltefosine to P. aeruginosa in vitro cultures decreased biofilm size. We will present data on miltefosine-mediated inhibition of P. aeruginosa PC degradation in liquid and biofilm cultures, and studies on PC-catabolite stimulation of P. aeruginosa biofilm formation. Introduction: Although eradication of non-mucoid strains of Pseudomonas aeruginosa (P.aer) is possible after first isolation, it is believed that mucoid P.aer strains are not eradicated. We hypothesised that with modern and aggressive antibiotic regimes, in particular nebulised antibiotics, this was no longer true.

Method: A retrospective chart review of all paediatric patients at one UK CF centre between July 1999 and July 2008 with at least one year follow up after first isolation of mucoid P.aer.

Results: Of 537 patients on the CF database, 138 (26%) isolated mucoid P.aer at least once from BAL, sputum or cough swab, and 116/138 had the minimum dataset. Sixty-seven of 116 patients (58%) cleared mucoid P.aer for >1 year; 38/67 patients (57%) remained clear at the last available culture (median 30, range 2-106 clear cultures, and median 55, 12-103 months clear.)

There was no difference in gender, genotype or age at first mucoid P.aer isolation between the group who cleared and those who did not, or those patients who remained clear of mucoid P.aer and those patients who re-cultured the organism after a year or more. There were no differences in FEV 1 between patients who cleared or did not clear, and patients who remained clear and those who did not, either before or after mucoid P.aer acquisition and before or after clearance.

Aspergillus fumigatus was found in a significantly higher number of patients who remained clear of mucoid P.aer (21/38) compared to those who became re-infected (7/29,p<0.05) .

Conclusions: Contrary to the literature, more than half of patients cleared mucoid P.aer for more than one year and one third achieved prolonged clearance, suggesting that aiming for eradication of mucoid P.aer is a realistic goal. The relationship between Aspergillus fumigatus and remaining clear of mucoid P.aer warrants further study. Dynamic microbial communities typify the airways of cystic fibrosis (CF) patients. Culture-independent approaches have been used to characterize the microbial constituents of such communities because it is thought that a more inclusive perspective of the lower airway microbiology is afforded by obviating the requirement for laboratory cultivation. A large proportion of the organisms reported in these studies are represented by bacterial species not identified by using the standard protocol for CF sputum microbiology, which has led to the notion that much of the microbiota present in CF airways is uncultivable. To better define the composition of the cultured airway microbiome of CF patients we used a more complete cultivation practice, which included several non-standard mediums as well as anaerobic growth conditions. We generated a strain collection of greater than 2000 isolates, recovered from hundreds of sputum specimens. Bacterial species identification was done by 16S rRNA sequence and revealed the presence of greater than 130 bacterial species from 40 different genera, many of which have only been detected by using culture-independent means. Additionally, several bacterial species were recovered that have not previously been described. In an attempt to evaluate the efficacy of the cultivation procedure we used to characterize the airway microbiome, a novel method was developed to discern the microbial community profiles recovered on twelve different culture mediums cultivated under two growth conditions (5% carbon dioxide and anaerobic). The culture-enrichment Terminal Restriction Fragment Length Polymorphism technique indicated that even our extensive collection of organisms underestimates the true magnitude of the cultured airway microbiome. The fusion of a culture-dependent and culture-independent approach allowed for a more comprehensive perspective of the lower airway microbiology by significantly increasing the sensitivity of either technique alone. We show that although each patient has a unique consortium of organisms, it is possible to culture the vast majority of organisms from CF sputum.

There is significant inter-individual variation in the rate of decline in lung function among patients with cystic fibrosis (CF), even among those with the same genotype. The differences may relate to the composition of the entire microbial community, not only those few bacterial species detected by conventional aerobic cultures. We performed ecological analyses on the full complement of throat swab microbes detected with culture-independent methods on 45 children with CF, 8 of whom had positive cultures for P. aeruginosa (PA). Microbial communities were more similar among homozygotes for dF508 than among those of patients with other genotypes. Fewer bacterial species were present in older patients, those on antibiotics, and in those with PA. The bacterial communities of different patients with PA had a smaller proportion of taxa in common than did those of patients without PA. Correspondence analysis showed that the PA-containing communities did not cluster with those of PA negative communities nor with each other (Fig.1 ). This implies that the community shift in composition might be either idiosyncratic or due to selection for a community metagenome that can be achieved by different sets of taxa. We conclude that respiratory microbial communities become smaller and their composition shifts significantly as patients age and acquire P. aeruginosa.

To identify major sources of variation in the dataset, we used correspondence analysis (a technique for identifying potential relations between variables when there are no a priori expectations as to the nature of those relations), which revealed that most bacterial communities grouped together (black dots), except those from individuals with culture-proven P. aeruginosa (gray dots). The airway microbiome in cystic fibrosis (CF) patients is complex and varies between patients and with pulmonary treatments. The clinical significance of these complex bacterial communities has not been established, and effective methods to monitor complex bacterial communities are needed. Anaerobic bacteria, in particular, have been associated with pulmonary exacerbation in CF patients, but little is known about the distribution of these organisms within the CF population.

Objective: To determine the prevalence of five selected anaerobic bacteria.

Approach: We are using quantitative PCR (qPCR) as an efficient way to monitor bacterial load and eight pre-specified organisms in standard of care respiratory tract samples.

Results: We have examined 165 sputum samples and 115 throat swabs from 116 patients over one year. Anaerobic bacteria were detected in the majority of all patients (72-91%). Further there is significant correlation between specific species of bacteria.

Conclusions: Quantitative PCR allows for detection of anaerobic bacteria in samples collected for standard of care microbiological monitoring. Anaerobic bacteria are present in a majority of samples obtained from CF patients.

This work was supported by the Cystic Fibrosis Foundation. Introduction: In adults with cystic fibrosis (CF), lung infections are typically polymicrobial and interactions between the different bacteria may be important contributors to CF lung disease. Anaerobic bacteria, especially Prevotella sp., have been detected in the sputum of adult CF patients and we showed that P. aeruginosa infection is associated with higher incidence of anaerobes.

Aim: Here we assess 1) the frequency of anaerobic bacteria in pediatric bronchoalveolar lavage (BAL) samples 2) their antimicrobial susceptibility and 3) serologic evidence of Prevotella infection.

Methods: Anaerobic bacteria were detected by complementary methods: Culture was performed under aerobic and anaerobic conditions using selective agars with viable bacteria counted and identified by sequencing 16S ribosomal RNA genes. Molecular detection of bacteria was done by extracting total bacterial DNA followed by terminal-restriction fragment length polymorphism (T-RFLP) analysis of PCR amplified 16S rRNA genes. Antimicrobial susceptibility was determined using E-tests®. Antibody response was examined by ELISA using whole cell bacterial lysates of P. aeruginosa and Prevotella melaninogenica.

Results: Bronchoalveolar lavage samples obtained from 33 CF infants and children (mean age 9.7 ± 0.8 yrs.) were examined. No bacteria were detected in 5/33 (15%) and 7/33 (21%) of these samples processed by culture and T-RFLP, respectively. The most frequent aerobes were S. aureus (13 and 10 samples) and P. aeruginosa (7 and 9 samples) detected by culture and T-RFLP, respectively. Anaerobically growing bacteria were cultured in 13/33 (39%) samples with Prevotella (4 samples) and Propionibacterium (3 samples) species predominating. T-RFLP detected anaerobic bacteria in 14/33 (42%) samples with Veillonella (9) and Prevotella (6 samples) predominating. The anaerobes cultured were generally resistant to tobramycin and metronidazole and exhibited species-specific resistance to clindamycin and oxacillin. All isolates were susceptible to meropenem and piperacillin/tazobactam. Serum samples were obtained from 38 pediatric CF patients: 18 with chronic, 10 with intermittent and 10 without P. aeruginosa infection, based on routine cultures. Average IgG titers to P. melaninogenica increased in parallel to P. aeruginosa titers for the 3 groups. Antibody titers for both P. aeruginosa and P. melaninogenica were significantly higher in the chronically infected than in the P. aeruginosa negative group.

Discussion: Anaerobic bacteria were detectable by molecular and culture methods in pediatric BAL samples. These bacteria show a high resistance to antibiotics used for treatment of CF exacerbations and those frequently used to treat anaerobic infections. Raised specific IgG levels suggest that P. aeruginosa and Prevotella melaninogenica infection may occur over a similar time frame. These data indicate that anaerobes occur early in CF lung disease and, if they are important in CF pathogenesis, would require adjustment of current antibiotic treatment regimens.

Supported by grants from Health and Social Care Research and Development, Public Health Agency, Northern Ireland, the Cystic Fibrosis Foundation and NHLBI. Bacterial lung infections are a major cause of morbidity and mortality in cystic fibrosis (CF). Traditional, culture-based diagnostic analyses of CF sputum have focused on a relatively small number of species, including the Burkholderia cepacia complex, Haemophilus influenzae, Staphylococcus aureus, Stenotrophomonas maltophilia and in particular, Pseudomonas aeruginosa. However, the recent application of culture-independent molecular analyses has indicated the presence of a more diverse range of species. Prevalent amongst these are anaerobic species such as members of the genera Veillonella and Prevotella, traditionally considered to represent contamination derived from oral flora.

It has been shown that a significant proportion of the bacteria colonizing the CF lower airways are growing anaerobically. We therefore hypothesised that anaerobic species detected in sputum samples, particularly those associated with the oral cavity, are representative of genuine colonisation of the CF lung rather than sample contamination.

Sputum samples were obtained from adult CF patients and frozen at -80 o C. Total DNA was extracted from a portion of the sample and 16S rRNA gene clone sequence analysis was performed. Samples in which members of the genera Streptococcus, Veillonella and Prevotella were identified were selected for further analysis. Interior regions of sputum boluses were sectioned to a thickness of 20 µm using a cryostat. One mm of sputum was discarded from each exterior face of sections to exclude salivary material. Sectioned sputa were analysed using a multi-labelled Fluorescent in-situ Hybridization (FISH) protocol, with probes specific to Pseudomonas aeruginosa, Streptococcus spp., Veillonella spp., and Prevotella spp. In addition, a bacterial DAPI stain was performed.

Initial data obtained through FISH analysis revealed Streptococcus spp., Veillonella spp., and Prevotella spp. were closely associated with clusters of P. aeruginosa throughout internal sections of sputum samples. Due to the location of these species, the data presented here suggests that species traditionally viewed as oral contaminants are in fact colonizing the CF lower airways. As such, the clinical significance of their presence must be considered. Introduction: Respiratory viral infections can lead to exacerbations of cystic fibrosis (CF) lung disease. Polymerase chain reaction (PCR) methods applied to nasal and/or throat swabs can be used for the rapid detection of common respiratory viruses. The majority of adults with CF produce sputum during infective exacerbations. We investigated whether sputum from CF patients is a suitable medium for the diagnosis of respiratory viral infections in patients with CF by a PCR method, by comparing results of PCR tests from paired sputum and nasal swab samples collected from patients with symptoms of a viral respiratory illness.

Methods: We prospectively collected paired nasal swab and sputum samples from patients attending the Manchester Adult CF Centre who had symptoms of a respiratory viral illness during the winter/spring season of 2008/2009. The nasal swab samples were transported to the laboratory in viral transpost media. The samples were investigated using a PCR method to detect common respiratory viruses (Rhinovirus, Influenza A, Influenza B, Parainfluenza type 1, Parainfluenza type 2, Parainfluenza type 3, Adenovirus, Respiratory Syncytial Virus, and Metapneumovirus). Permission for the study was granted by the local medical ethics committee.

Results: Paired samples were analysed from 42 patients. There were 20 (47.6%) positive results: 10x Rhinovirus, 3x Influenza A, 3x Influenza B, 2x Parainfluenza type 3, 1x Respiratory Syncytial Virus, 1x Metapneumovirus. Of the 20 positive results, all (100%) of sputum samples were positive, whilst 15 (75%) of nasal swab samples were positive.

Discussion: Respiratory viral infections in adults with CF can be confirmed by PCR for common respiratory viruses applied to sputum samples from CF patients. In our clinical practice, the specificity of PCR is greater using sputum samples than nasal swab samples. In vitro experiments demonstrate that genomic mutations can lead to increased persister formation. We hypothesized that persisters enable establishment of chronic infections despite aggressive antibiotic therapy and that periodic treatment of a patient over time will select for mutations leading to increased persister formation. To test these hypotheses we obtained longitudinal clonal clinical isolates of P. aeruginosa from chronically infected cystic fibrosis (CF) patients. Thirty-five of these isolates obtained from a single patient with CF were evaluated for antibiotic tolerance during log phase growth, stationary phase, and biofilm growth. During stationary phase, between 1x10 -5 and 1x10 -4 percent of the population of the early isolate are drug-tolerant persisters, while between 0.01 and 0.1% of the population of the last isolate are persisters. Thus the last isolate exhibits a 100-1000 fold increase in persister cell formation at stationary phase compared to the early parent strain. During exponential phase, between 0.1 and 0.5% of the early isolate population formed persisters while between 5-10% of the last isolate population tolerated this antibiotic, indicating at least a 10-fold increase in persister formation in the last isolate. In biofilms, the last isolate formed between 0.02-0.05% persisters while the parent strain formed no persisters at any of the antibiotic concentrations tested. This is a surprising result given that the last isolate is partially defective for in vitro biofilm formation. The early isolate forms bioflms with approximately 1x10 7 CFU/peg while the last isolate forms biofilms of approximately 1x10 5 CFU/peg. For all antibiotics tested the last isolate exhibited higher resistance. To distinguish resistance from the persister phenotype we knocked out the MexXY pump from the last isolate, which was upregulated due to a mutation in mexZ not present in the parent strain. This resulted in parental MIC levels in the last isolate without affecting persister levels. In addition to the isolates from this patient, clonal pairs from 13 other patients were assessed for their ability to form persisters at stationary phase. The last isolate from 10 of these 13 pairs formed at least 10-fold more persisters at stationary phase compared to the paired early isolate. Moreover, in 5 of these 10 pairs, there was no significant increase in antibiotic resistance between the early and the late isolate indicating that increased survival was due to increased persister cell formation and not classical antibiotic resistance. Together, these results suggest that increased persister formation by P. aeruginosa is an important adaptation in chronic infection of the CF airway. Support: L.R Mulcahy is supported by a Cystic Fibrosis Foundation Postdoctoral Fellowship (MULCAH08F0). Strain sharing was supported by a CFF Clinical Antimicrobial Toolkit Award to J.L. Burns. Objectives: This study investigates the prevalence and antimicrobial resistance of MRSA isolates in the Northern Ireland adult CF population, lung function changes associated with MRSA positive sputum and antibiotic treatment, antibiotic choices, the length of treatment and the efficacy of treatment.

Method: A retrospective study was carried out on all patients with CF attending the NI Regional adult CF centre, Belfast City Hospital, with a history of MRSA between 1999 and February 2009. Case notes were used to ascertain FEV1 three months prior to infection, at date of initial sputum culture positive MRSA, at the end of antibiotic treatment, and three months post-infection. Information on the antibiotics used and duration of treatment were also obtained.

Results: Twenty eight adult patients (mean age 25 (7) years) were included in the study. FEV1 (%predicted) at initial positive sputum result was [64 (24)]. The majority of infections (82%) were treated with 6 weeks of rifampicin dose 300mg po bd and fusidic acid 500mg po bd and the remainder with combination therapy based on antimicrobial sensitivity and reported side effects. FEV1 (%predicted) showed a small but significant improvement after antibiotics for first MRSA infection only [7(15)%], p=0.04, (95% CI 0-13%). Overall 26 of the 28 patients remained MRSA sputum culture negative for six or more months following antibiotic treatment. Twentythree patients were treated with rifampicin and fusidic acid, of whom 21 had successful eradication of MRSA. Linezolid was used in 5 infections involving four patients with recurrent MRSA infections and proved ineffective in 4 of these.

Conclusions: This study demonstrates that FEV1 improves significantly following antibiotic treatment for first isolation of MRSA from sputum culture in CF. Rifampicin and fusidic acid for six weeks are effective firstline agents.

Reference : Background: Infection control is important for patients with cystic fibrosis (CF). We are a large regional centre with over 300 adults and operate a strict infection control policy. This includes specific cohort clinics based on sputum microbiology; isolation within clinics so that members of the MDT rotate around the patient; a dedicated in-patient facility with single rooms; patients with Burkholderia cepacia complex (Bcc) infection are housed on a separate general respiratory ward. Patients with MRSA infection are barrier-nursed and inpatients are housed in single rooms with ensuite facilities. We operate a policy of eradication of MRSA. Sputum is sent for culture (including MRSA) at each clinic visit and during each inpatient admission.

The most common healthcare associated strains of MRSA in the UK are EMRSA-15 and EMRSA-16.

Aim: Review the results of strain typing of MRSA isolates and epidemiological data to identify if any cases of MRSA cross-infection occur within our centre.

Method: Patients with a positive sputum culture for MRSA were identified from our microbiological database. Case notes, epidemiological data and results of strain typing via DNA fingerprinting using pulsed field gel electrophoresis (PFGE) and phage typing from a national reference laboratory (Laboratory of HealthCare Associated Infection, Colindale) were reviewed.

Results: Thirty-seven of 469 (8%) of patients who have attended the unit since 1998 have isolated MRSA in their sputum. Strain typing results are available on 24 of these isolates. Eleven (29.7%) patients had isolated MRSA prior to transferring from their paediatric unit. Of the 26 adults who acquired MRSA whilst attending our unit there were identifiable risk factors in 13 (50%). These included; 4 patients transferred from other adult units; 1 patient had had an admission at a district general hospital; 3 were either health care workers or married to one; 2 patients are married to each other; 1 had a husband with a chronic medical condition who is positive for MRSA; and 2 patients were positive for Bcc and were nursed on a general respiratory ward at our centre. Fifteen different pulsotypes of MRSA were identified. Twelve were identified as sub-types of EMRSA-15. Of these, 4 were indistinguishable (EMRSA-15 new variant B3). Five were identified as EMRSA-16 of which 2 were indistinguishable (EMRSA-16 variant A1). The other 7 isolates were representative of more minor clones of HA-MRSA circulating in the UK.

Of the patients with identical strains, a review of epidemiological data including the dates of the initial positive sputum culture, dates of admission and dates of clinic attendance showed that cross-infection between patients was extremely unlikely.

Conclusion: Typing and epidemiological data confirms there is no evidence of cross-infection with MRSA within our large CF centre. The results validate the place of rigorous infection control measures as part of optimal CF care. Strict adherence can prevent direct patient-patient transmission of MRSA.

Vosahlikova, S. 1 ; Omelka, M. 2 ; Drevinek, P. 1 

In our previous study, approximately 30% of Czech patients with cystic fibrosis (CF) were confirmed as being infected with epidemic strain of Burkholderia cenocepacia ST-32. Although anti-epidemic measures and patients' segregation based on their microbiological status were put in place in early 2000s, new cases of infection still appear. Their source is unknown; however, these cases may still be linked to the past epidemic outbreak if ineffective separation of CF patients or leaks in the anti-epidemic measures in the hospital take place. The aim of this study was to track potential routes of ST-32 infection among CF patients at a Prague CF Center by retrospective evaluation of their contacts using hospital records, in combination with molecular typing of B. cenocepacia.

We applied a statistical procedure for a retrospective evaluation of potential infectious contacts that was independent on assumptions about incubation periods, infectivity or other epidemiological parameters. The analysis was based on the archived records from health insurance bills from 364 CF patients that were treated at the Prague CF Center over the period of 1999 to 2007. Every record from the billing system contained unique patient's identifier, the date of hospital visit, and ID for hospital unit (HU; the lowest distinguishable part of hospital in the billing system). Comparing dates of HU visits, we assessed the total number of all potential patientpatient contacts which may have occurred during the study period, a number of risk contacts; i.e. a situation where a positive and a negative patient came to the same HU at the same day, and a number of CF patients with newly diagnosed B. cenocepacia infection whose acquisition may have been a direct consequence of the risk contact. Bacterial genotyping was performed with PFGE and MLST.

Over the period of nine years, we assessed 34491 contacts which occurred at 176 HUs based on billing data. Approximately 18% (6180) of them were marked as risk contacts and detected at 39 HUs. However, no patient eventually became infected after visiting any of 33 out of 39 HUs. Remaining 6 HUs were considered sites of potential transmission since 14 CF patients who visited these HUs the same day as a B. cenocepacia positive patient got infected with an identical strain (ST-32). These 6 HUs represented 37% (2297) of all risk contacts and the 14 patients represented 50% of all patients who were diagnosed with B. cenocepacia ST-32 infection over the study period (no epidemiological link was detected in the remaining 50% of patients). Based on the number of infected CF patients, a lung function test laboratory was identified as the highest risk HU where risk contacts were detected for 6 out of 14 CF patients.

Our study highlights the importance of regular re-assessment of antiepidemic measures in every epidemiologically relevant HU including those which might not be an inherent part of the CF center, and the need of continual staff education in order to prevent true risk contacts from happening. Objective: Chronic azithromycin therapy is recommended for patients with cystic fibrosis (CF) who are chronically colonized with Pseudomonas aeruginosa and are greater than 6 years of age. Macrolide antibiotics are widely reported to cause prolongation of the QT interval, which may lead to Torsades de pointes. Although mainly associated with the use of erythromycin and clarithromycin, there is growing evidence that azithromycin has the potential to prolong the QT interval as well, and there are several reports of torsades de pointes associated with azithromycin use in adults. This pilot study was performed to assess the incidence of prolonged QT intervals in pediatric CF patients receiving chronic azithromycin therapy.

Methods: We identified patients on chronic azithromycin therapy (defined as greater than 2 months duration) from ages 1 to 17 years and obtained electrocardiograms (EKGs) on the patients during their clinic visits. EKGs were interpreted by a board certified pediatric cardiologist (HL) and the corrected QT interval (QTc)was calculated using Bazett's formula. A QTc ≤440 ms was interpreted as normal, QTc 441-460 ms was interpreted as borderline, and QTc >460 ms was interpreted as prolonged.

Results: A total of 37 patients on chronic azithromycin therapy were identified. The mean(±SD) QTc was 417±19 ms. Average QTc was longer in patients homozygous for ∆F508 (424±19 ms; n=22) than in patients with one (n=14) or no (n=1) ∆F508 mutations (406±15 ms; p=0.005). No patients were identified as having prolonged QTc >460 ms; however, 4 patients had QTc >440 ms. Of interest, all of the patients with borderline QTc intervals were homozygous for ∆F508. The prevalence of QTc prolongation among ∆F508 homozygous patients (18%) was similar to the reported 25% incidence in children treated with clarithromycin.

Conclusions: Chronic azithromycin therapy is associated with prolongation of the QTc in children with cystic fibrosis at approximately the same frequency as reported with clarithromycin. Since patients with CF receive azithromycin for prolonged durations, it may be prudent to screen patients for prolongation of the QTc prior to and after initiation of chronic azithromycin therapy. We describe the characteristics of the study population and the conduct of a randomized trial of young children with cystic fibrosis (CF), at first isolation of Pseudomonas aeruginosa (Pa) from their airways. The specific objectives of the trial were to compare the clinical and microbiological efficacy and safety of quarterly cycled antibiotic therapy versus quarterly culture-based therapy, initiated at the time of initial Pa acquisition.

Methods: Randomized multicenter trial in CF patients ages 1-12 years at first isolation of Pa from a respiratory culture. Using a factorial design, trial participants were assigned for 18 months to either anti-pseudomonal treatment on a scheduled quarterly basis (cycled therapy) or based on recov-ery of Pa from quarterly respiratory cultures (culture-based therapy). Treatments consisted of inhaled tobramycin (300 mg BID) for 28 days, combined with either oral ciprofloxacin (15-20 mg/kg BID) or oral placebo for 14 days. The study was approved by the IRB at each participating institution. The primary endpoints of the trial were 1) time to pulmonary exacerbation requiring IV antibiotics or hospitalization for respiratory symptoms, and 2) the proportion of patients with new Pa-positive respiratory cultures during the study.

Data: Fifty seven clinical centers throughout the US participated and the enrollment goal of 300 participants was met in 2.5 years. Between December 20, 2004 and February 7, 2008, 336 patients have been screened and of those, 308 were eligible for enrollment. Of the 308 eligible patients, 304 were equally randomized to either cycled (n=152) or culture-based therapy (n=152). Of those, 24 patients (8%) have withdrawn during the study follow up. Among participants who completed the study, there was only a very negligible number of missed visits (12/2155, 0.5%). Ninety patients were ages 1-3 years, 84 ages >3-6 years, and 130 ages older than 6 years. Comparable numbers of males and females participated, 49% of patients were Delta F508 homozygous, and 95% were pancreatic insufficient. Approximately half of the patients were exposed to anti-pseudomonal antibiotics within 6 months prior to randomization.

Conclusions: Considering the duration of the study, a negligible number of patients withdrew prior to termination of the study and had incomplete follow-up (<10%). Among patients who completed the study, the follow up was nearly complete with only 0.5% of missed visits. The age distribution was uniform across age categories, and baseline characteristics were uniform across age groups. Baseline characteristics stratified by group assignment and details regarding the quality of the study conduct will be presented.

Acknowledgements: Supported in part by the CFF grant number EPIC0K0, NHLBI and NIDDK grant number U01-HL080310, NCRR grant number ULI-RR2501401, Novartis Pharmaceutical Corp. (inhaled tobramycin), Bayer Healthcare AG (oral ciprofloxacin and oral placebo), and PARI Respiratory Equipment Inc (compressors and nebulizers). 

Introduction and aims: We have previously shown that photodynamic antimicrobial chemotherapy (PACT) can be used to achieve high rates of kill of clinical cystic fibrosis (CF) Pseudomonas aeruginosa isolates growing planktonically and in biofilm. The aim of this study was to determine if exposure of clinical P. aeruginosa isolates to sub-lethal doses of PACT could result in decreased susceptibility to 1) previously lethal PACT and 2) antibiotic treatment.

Methods: Planktonic cultures of P. aeruginosa PAO1 and clinical isolates (n=4), cultured from the sputum of CF patients, were exposed to sublethal methylene blue (MB) and meso-tetra (N-methyl-4-pyridyl) porphine tetra tosylate (TMP)-mediated PACT over a 72 hour period. The susceptibility of surviving bacteria to a range of antibiotics (tobramycin, ceftazidime, meropenem, piperacillin/tazobactam) currently used in the treatment of CF pulmonary infection was then determined using E-test® strips and compared with the susceptibility of untreated controls. Surviving bacteria were also treated with previously lethal photosensitizer-light combinations and the percentage kill determined.

Results: The isolates were in general more susceptible to TMP than MB-mediated PACT with higher percentage kills apparent with TMP for all isolates tested. For antibiotic sensitivity results, the percentage of antibiotic/isolate combinations where the minimum inhibitory concentration (MIC) of test samples deviated from that of the untreated controls (exposed to neither light nor photosensitizer) were calculated (Table) . Exposure to sublethal PACT did not affect the subsequent antimicrobial susceptibility of the isolates with 80% of MICs for treated isolates either identical or within one doubling dilution of the MICs for untreated controls. Furthermore, exposure to sub-lethal PACT did not result in any change in the categorization of isolates as susceptible, resistant or intermediate according to the Clinical and Laboratory Standards Institute breakpoints. In addition, exposure to sublethal PACT resulted in no significant decrease in susceptibility to previously lethal PACT.

Conclusion: Sub-lethal PACT exposure does not affect the susceptibility of clinical P. aeruginosa isolates to either previously lethal PACT or antibiotics currently used in the treatment of CF pulmonary infection. Therefore, PACT remains a potential option for treatment of P. aeruginosa CF pulmonary infection. Introduction and aims: Antimicrobial susceptibility testing is valuable for directing the choice of antibiotics for treatment of pulmonary infection in individual CF patients and for monitoring the prevalence of resistant strains within a CF centre. The aim of this study was to determine the prevalence of resistance amongst Pseudomonas aeruginosa (PA) isolates, cultured from CF patients attending the Adult CF centre in Belfast, to a range of antibiotics widely used in the treatment of PA respiratory infection.

Methods: The susceptibility of PA isolates to 6 antibiotics was determined using E-test ® strips with minimum inhibitory concentrations determined after incubation for 24 hours at 37 ᭺ C. Isolates were classified as susceptible, intermediate or resistant to each antibiotic according to the Clinical and Laboratory Standards Institute parenteral breakpoints. Of the 76 isolates, 51 were cultured from patients participating in an exacerbation study. These isolates were collected when patients were stable and prior to and after antibiotic treatment for an acute infective exacerbation. The remaining 25 isolates were Liverpool epidemic strains (LES) which are thought to be associated with a poorer outcome.

Results: Thirty-four isolates (45%) were susceptible to all six antibiotics with no isolates resistant to all six agents tested. The majority of patients with an infective exacerbation were treated with a combination of tobramycin and either ceftazidime or piperacillin/tazobactam. Resistance to tobramycin occurred in only 2.6% of the isolates while greater resistance to ceftazidime (39.5%) and piperacillin/tazobactam (43.4%) was observed. Notably, there was generally no difference in the antibiotic susceptibility pattern of isolates cultured from the same patients, pre-and post-antibiotic therapy. Despite its use for chronic maintenance therapy in approximately 80% of patients, resistance to colistin occurred in only 1.3% of the isolates. Resistance to meropenem (18.4%) and aztreonam (31.6%) was observed. Lastly, no differences were apparent in the sensitivity of the exacerbation and the LES isolates for any of the antibiotics tested.

Conclusion: This study has shown that current treatment strategies within the centre are not leading to the development of widespread antibiotic resistance. Such snapshot surveys to identify the development of resistant strains within CF centres may improve centre performance by informing prescribing policies.

Supported by Gilead Sciences, Inc. 

Objectives: MRSA is being increasingly cultured from the respiratory tract of patients with CF although the contribution of MRSA to declining lung function remains unclear. This study characterized and compared MRSA isolates cultured from paediatric CF patients in Northern Ireland and the USA to study the epidemiology of CF-associated MRSA.

Methods: Patients were recruited from the Belfast Paediatric CF centre (BPCF) (n=21) and from the Paediatric CF centre at the University of North Carolina (UNC) (n=7). MRSA isolates were characterized by: 1) SCCmec complex typing; 2) pulsed field gel electrophoresis (PFGE) and subsequent analysis of banding patterns by UPGMA and Dice coefficients; 3) spa typing 4) agr typing (as determined by PCR and ScaI restriction fragment length polymorphism (RFLP)) and 5) PCR detection of the PVL gene. The ability of each isolate to form a biofilm under anaerobic and aerobic conditions was determined by growth in a 96-well plate and staining with crystal violet.

Results: Twenty-two isolates (cultured from cough swabs and sputum) from 21 BPCF patients and 11 isolates (cultured from bronchoalveolar lavage fluid (BALF) samples) from 7 UNC patients were compared. BPCF isolates were mostly SCCmec IV (21/22 isolates) whilst UNC isolates were SCCmec II (n=8/11) or IV (n=3/11). By PFGE, 21/22 BPCF isolates, formed a closely related cluster of >60% similarity, spa type t032, similar to EMRSA-15. All of these isolates were agr type I and none were PVL positive. Nine of 11 UNC isolates and the remaining BPCF isolate, also formed a cluster of >70% similarity to each other, with spa types and PFGE patterns related to the New York/Japan MRSA clone and were agr type II. The remaining 2 UNC isolates were PVL positive, resembling the USA-300 MRSA clone. Multiple MRSA isolates from one sputum or BALF sample shared the same PFGE pattern, suggesting that patients were colonized with one MRSA strain. BPCF isolates included isolates from 2 sibling pairs; one pair was identical, with exact banding pattern matches, while the other pair had a one-band difference. Furthermore, duplicate isolates from one patient, obtained 18 months apart, were similar, but not identical (a 2-band difference). Irrespective of agr type and source, all isolates formed weak biofilms with no difference observed in biofilm formation under aerobic and anaerobic conditions. Conclusion: UNC and BPCF MRSA isolates differed from each other and likely reflected the predominant MRSA strain types circulating in their respective countries. Molecular epidemiological studies to investigate the potential impact of the differing MRSA strains on clinical outcome within the CF population warrant further investigation. Pseudomonas aeruginosa is a gram-negative opportunistic pathogen that causes chronic infection in the lungs of cystic fibrosis (CF) patients. P. aeruginosa has been shown to secrete a soluble protein, termed CFTR inhibitory factor (Cif), which affects the trafficking of select ABC transporters; including CFTR and P-gp. The result is a rapid mislocalization of the transporter to endosomes, followed by degradation. Cif's effect on CFTR causes inhibition of chloride currents across the membranes of treated cells. Since the majority of CF patients are colonized by P. aeruginosa, Cif represents a potential roadblock to the effectiveness of CFTR targeted therapies in a clinical setting. Cif is a member of the α/β hydrolase superfamily of proteins, and is an epoxide hydrolase (EH). Mammalian EHs can function as regulators of blood pressure and inflammation, while bacterial EHs have mostly been characterized for their ability to detoxify environmental and industrial compounds. This is the first example of an EH functioning as a virulence factor.

We have determined the X-ray crystal structure of Cif to 1.8 Å. The enzymatic mechanism utilizes a catalytic triad of residues Asp129, Glu153, and His297. His269, previously implicated as an active site residue, is located at the mouth of the putative substrate access tunnel to the active site. This mutation does not impair enzymatic function with low molecular weight substrates in vitro, but does abrogate all measureable effect in vivo with airway epithelial cells, classifying it as a potential K m mutant. The structure provides a basis for investigating the role of the active site in Cif activity. The intracellular signaling molecule, cyclic-di-GMP, has been shown to influence bacterial behaviors, including motility and biofilm formation. Given that Pseudomonas aeruginosa is able to chronically colonize the lungs of cystic fibrosis (CF) patients leading to an increase in morbidity and mortality, we would like to better understand the mechanisms by which the c-di-GMP signaling pathway impacts surface behaviors of this bacterium. Recently, we reported the characterization of the bifA gene, a regulator of surface-associated behaviors in P. aeruginosa. The bifA gene encodes a c-di-GMP phosphodiesterase and the ∆bifA mutant exhibits increased cellular pools of c-di-GMP, forms hyper-biofilms and is unable to swarm. The hyper-biofilm phenotype results from an increased synthesis of a polysaccharide produced by the pel locus. However, this increased polysaccharide production in the ∆bifA mutant has only a subtle inhibitory effect on swarming motility. In this study, we sought to understand why the ∆bifA mutant is unable to swarm and to elucidate the potential mechanism(s) by which c-di-GMP impacts swarming motility.

Methods: We performed a transposon mutagenesis using the Mariner transposon in the ∆bifA mutant background and screened for mutants that recovered the ability to swarm. Selected mutants were characterized in detail using molecular and biochemical studies.

Results: We isolated several mutants containing mutations in the pilY1 gene, previously implicated in twitching motility -another form of surface motility used by P. aeruginosa. The PilY1 protein is homologous to the PilC protein of pathogenic Neisseria species. PilC is required for pilus formation and adhesion to human epithelial cells. In P. aeruginosa, mutation of the pilY1 gene in the ∆bifA mutant background not only suppresses the swarming defect but also leads to a reduction in pel polysaccharide production and hence an abrogation of the hyper-biofilm phenotype. We are currently investigating the potential link between pilY1 gene function and the cdi-GMP signaling pathway.

We have identified a potentially novel component in the c-di-GMP signaling pathway that is able to influence all three surface-associated behaviors of P. aeruginosa -twitching, swarming, and biofilm formation and thus might impact the ability of this bacterium to colonize niches such as the CF lung environment.

Arets, B.; Bonestroo, H.J.; de Winter-de Groot, K.; van der Ent, K.

Patients with cystic fibrosis (CF) show an increased bacterial colonization of both upper airways (UAW) and lower airways (LAW). The relation between cultures from these two locations is unknown, as is the relation between positive cultures and clinical characteristics. The aim of this study was to analyze 1) differences between the results of simultaneously taken UAW and LAW cultures in children with CF and 2) differences in clinical characteristics between patients with positive versus negative LAW and UAW cultures.

Methods: Bacteriological and clinical data from 157 young CF patients were collected during annual check-up from the database of the CF Center Utrecht. The number of positive nasal (UAW) and sputum (LAW) cultures for specific microorganisms and the correspondence between these results and clinical characteristics were analyzed.

Results: Bacterial growth was found in 79.6% of LAW cultures versus 43.9% of UAW cultures (p<0.001). P. aeruginosa (PA) was found in 34.4% vs 11.5% (p<0.001) and S. aureus (SA) in 50.3% vs 26.8%, respectively (p<0.001) ( Table 1 ). Compared to patients with a negative LAW culture, patients with a positive culture were significantly older (9.8 vs 11.9 yrs, p=0.002), and had more complaints of cough (65.6% vs 85.6%, p=0.009), sputum production (46.7% vs 73.6%, p=0.004) and nasal polyposis (34.4% vs 56.8% p=0.023). Patients with a positive UAW culture had more nasal discharge compared to negative culture patients (50.7% vs 25.0%, p =0.001). Patients with a PA positive LAW culture were significantly older (12.9 vs 10.8 yrs, p=0.001), had a worse lung function (FEV1%pred 85.4% vs 92.5%, p=0.039) and more sputum production (81.0% vs 61.0%, p=0.009). Patients with SA positive UAW cultures had nasal discharge more often (52.0% vs 30.0%, p=0.011). In more than half the patients with a positive UAW and a negative LAW culture for PA, the LAW became positive within the next 3 months.

Conclusion: In this study, we found a difference between UAW and LAW cultures in patients with CF.

Besides, we found differences in clinical characteristics between patients with positive versus negative culture results. PA positive UAW cultures appeared to precede positive LAW cultures in a substantial part of patients, suggesting some kind of cross-infection between the UAW and LAW. Mycobacterium avium forms biofilms. Studies in vitro have suggested that M. avium in biofilm is resistant to antibiotics commonly used for treatment of the condition in humans. In experimental animals, the ability to form biofilm is associated with efficiency to establish disease. We hypothesize that biofilms can resist the host immune defense, and we studied whether macrophages can affect M.avium biofilm at different stages of evolution.

Methods: MAC 104 and A5 biofilm were established on glass, plastic and polarized human bronchial cells. After 14 days, THP-1 macrophages were added to the biofilm, and after 4 h the biofilm was plated for CFU and scanning EM. Macrophages were also stimulated with IFN-γ and TNF-α for 24 h before adding to the biofilm. In addition, biofilm was seeded and exposed to macrophages at days 1, 5 and 7 for 2 days. Viable bacteria were plated for CFU. Video and scanning electro microscopy were carried out. M. avium biofilm matrix was analyzed.

Results: Biofilms of neither MAC 104 nor A5 were affected by addition of resting macrophages or activated macrophages. Macrophages were "hyper-activated" according to EM, O2-and NO production data. However, macrophages had impaired ability to ingest and kill M. smegmatis. Contact with macrophages upon seeding of biofilm led to decrease in bacterial load. Biofilms older than 24 h, however, were not affected by macrophages. Presence of biofilm was associated with increased macrophage apoptosis. Several proteins specific for M.avium biofilm were identified.

Conclusion: M. avium biofilm is harmful to both resting and activated macrophages, inducing apoptosis. Macrophages exposed to biofilm are functionally impaired. Investigation of the effect of biofilm components on macrophages may provide information regarding therapy. Aim: We studied whether the sinuses may serve as a place for bacterial adaptation and focus for lung colonisation and infections. It has been shown that patients who have had a lung transplant are recolonised in their lung grafts with the same clones as those that were cultured from their old lungs.

Methods: We have treated 54 cystic fibrosis (CF) patients (median age 17.5 years, range 6-50) for chronic sinusitis with Functional Endoscopic Sinus Surgery. We evaluated whether their sinuses serve as a bacterial reservoir, and measured IgG and IgA antibodies against P. aeruginosa (PA) sonicate and alginate in serum and saliva. Nineteen patients were chronically infected (13 with PA, 1 with Burkholderia, 4 with Achromobacter, 1 with Moraxella), 21 were intermittently colonised (19 with PA and 2 with other CF pathogens) and 14 have only been colonised once with PA during the most recent 2-year period.

Results: In chronically infected patients there was complete agreement between lung and sinus bacteriology and genotypes. In children intermittently colonised with PA there was agreement in 17/19 (89%) of the cases and in patients from whom we have only cultured PA once no PA was cultured from the sinuses. In chronically infected patients PA were organised in biofilm structures in the sinuses similar to those seen in sputum. In contrast to biofilms from the lungs, we found that biofilms from the sinuses were only surrounded by very few inflammatory cells. We measured an IgA response in saliva against PA sonicate and alginate that was 19 and 61 times higher than in serum indicating a local production in the airways. The IgA production in serum and saliva against PA was significantly higher than the IgG production (p<0.05) where no difference between saliva and serum was found.

Conclusions: The anti-phlogistic IgA probably prevents adhesion of PA to the epithelial surface of the sinuses and thereby also a PMN inflammatory response. Since PA escapes the systemic immune system because of adhesion to IgA in the mucus, the systemic IgG response is not activated and the biofilm growing PA persists. As a consequence PA adapts silently to the airways subsequently leading to colonisation and infection of the lungs with adapted bacteria, which are difficult to eradicate since they grow slowly, and are resistant to antibiotics. Chronic airway infection by Pseudomonas aeruginosa (PA) is a specific hallmark of cystic fibrosis (CF) lung disease. Airway secretions are routinely cultured for bacteria, but this approach is of limited sensitivity in younger, non sputum-producing patients. Therefore, non-invasive real-time tests would offer significant advantages in the surveillance, detection and subsequent early treatment of PA infection.

PA organisms are capable of releasing the toxic gas hydrogen cyanide (HCN) which contributes to epithelial damage. In recent work, sputum cyanide levels have been linked to PA density and lung function. However, HCN concentrations as detected by selective ion flow-tube mass spectrometry were highly variable and often close to the detection limit.

We developed an alternative analytical approach: A tunable diode laser absorption spectroscopy (TDLAS) system for measuring HCN in humid gas samples. With data acquisition times as short as 1 minute, we were able to demonstrate high precision and specificity with lower limits of detection around 1-2 ppb. In 100% of air samples from exsiccators in which clinical PA isolates had been cultured for 48hrs, we measured significant HCN production (approx. 400 ppb) by TDLAS. Comparable experiments for S. aureus, H. influenzae, E. coli, Stenotrophomonas, Achromobacter and Burkholderia species are under way.

Therefore, HCN may hold potential as an exhaled biomarker for PA infection in CF patients. We are planning to improve experimental sensitivity and then investigate breath samples from CF patients to establish the diagnostic value of exhaled HCN levels. 2008;180:5662) . PGP shares structural features with CXCR ligands (eg. IL-8) and independently drives pulmonary neutrophil influx. PGP and its acetylated form (Ac-PGP) are found in high levels in CF sputum vs. healthy and disease controls, and inhibition of MMP-9 blocks >80% of PGP generation. CF patients have high sputum levels of MMP-9 that lacks negative regulation, with activity exceeding that of children intubated with respiratory failure and Acute Lung Injury (ALI). When examining MMP-9 activity in intubated-mechanically ventilated children, we found that the highest levels were seen in patients with RSV-induced disease (12-fold increase, p<0.002, n=23 for children with RSV, and n=21 for children without RSV). These results suggest that RSV is a specific stimulus for MMP-9 expression. It has been postulated that viral infections are an important early life stimulus to initiate the progressive cascade of lung inflammation and damage seen in CF. Based on these observations, we tested the hypothesis that RSV infection induces MMP-9 expression and release from human airway cell monolayers. 16HBE cells and wtCFTR+ CFBE41o-cells were grown as polarized monolayers on permeable supports and infected with RSV at an MOI of 10. Cell lysates and media were assayed for MMP-9 expression and release from day 01 through day 12 post-infection. Our results indicate that RSV A2 induces MMP-9 expression in both cell types, with increased transcript levels in 16HBE cells of 5, 25, and 150-fold over non-infected controls at days 03, 06, and 09 (p<0.02 at all time points). Complementary Western blots and activity assays performed on cell media indicated that MMP-9 zymogen increased in a qualitatively similar fashion over the course of RSV infection, with delayed and modest induction of TIMP-1 expression at later time points (maximum of 4-fold over uninfected controls at day 09 post-infection). Comparisons of apical and basolateral MMP-9 release indicated that MMP-9 was released into both compartments, with apical > basolateral release at all time points tested. Control experiments with heat-killed RSV and non-replicating adenovirus exposure confirmed that induced MMP-9 expression was specific for live RSV virus. The results support the hypothesis that RSV infection is sufficient to induce MMP-9 expression in human airway cells. The release of MMP-9 to the apical compartment suggests that the high levels detected in airway secretions relative to non-RSV diseases represent MMP-9 derived from neutrophils and the epithelium. The detection of basolateral release suggests that epithelial-derived MMP-9 has access to extracellular matrix proteins and may contribute to the generation of PGP and damage to airway structural proteins.

Drevinek, P.; Vosahlikova, S.; Reitzova, H.; Cinek, O.

Highly sensitive detection of the Burkholderia cepacia complex (Bcc) organisms in sputum of cystic fibrosis (CF) patients has been made possible by introducing a nested PCR into the microbiological investigation scheme. It allowed diagnosing infection in its early stage, often when the culturebased analysis was still negative. However, such samples which had been found positive only by PCR could not be analyzed for epidemiological purposes any further since traditional typing methods required a pure culture. On the contrary, recently developed Multilocus Sequence Typing (MLST) which is based on DNA sequencing of seven housekeeping genes does not rule out its applicability in situations where a bacterial culture is not yet available.

To evaluate whether strain identification is possible directly from clinical material, we carried out a pilot study on 20 samples from 16 CF patients which differed by: (i) intensity of Bcc positivity ( To increase method sensitivity and make it comparable to PCR-based detection from sputa, we took an advantage of originally developed MLST scheme for the Bcc where primers had been designed separately for gene amplification and for sequencing. As the latter primers formed the inner pair, they were found suitable also for a second round of nested PCR.

We identified bacterial strain type (ST) for every patient, although gene amplification or sequence analysis failed in several individual reactions (ca. 3%). In such event, ultimate ST identification was achieved by examining another sample from the same patient. STs corresponded to Bcc species examined and the two previously indeterminate samples were assigned to Burkholderia contaminans. No interference with DNA of either human or other bacterial origin was detected.

This proof-of-principle study showed that strain identification of Bcc bacteria is feasible directly from sputum samples, and the method is particularly useful in situations where cultivation results remain negative due to a low bacterial load.

Supported by grants of Czech Ministry of Education MSM0021620812 and VZ642036405. Aim: compare 2 Pa antibiotic (AB) eradication regimens. Method: Cystic fibrosis (CF) children (0-18 years) with new isolation of Pa (sputum or cough swabs) were prospectively randomized to treatment with inhaled TOBI® (300 mg bid for 28 days) vs 3 months combination therapy with inhaled Colistineb® (2 milj U bid) + oral ciprofloxacin (10mg/kg tid)(CC). Monthly airway cultures were taken for 6 months, then 3 monthly. Primary outcome was Pa eradication at end of treatment. Secondary outcomes included sputum culture in the first 6 months defined as eradication if repeatedly Pa negative. Patients who failed the primary outcome were changed to the other treatment arm or received IV AB if clinically indicated. Other secondary outcomes were: Pa antibodies (Ab), IgG, lung function, weight and time to Pa relapse.

Results: Fifty patients have been included so far. Interim analysis on 32 patients is available (15 CC and 17 Tobi). For 8 patients it was the first Pa isolation ever. The CC group has a mean age (SD) of 8.3 (5.1) years with a mean (SD) FEV1 of 97 % pred (11.8), Pa Ab 0.79 (2.4) (AU) and IgG 8.8 (2.6 ) g/l. The TOBI group has a mean age of 9.5 (4.3) years with an FEV1 of 101 % (12) pred, Pa Ab of 1.38 (4.7) AU and IgG of 8.2 (2.4) g/l. Initial Pa clearance was achieved in 14/15 CC and 14/17 TOBI (p= 0.14). Eradication at 6 months follow-up was documented in only 7 patients in each treatment arm (p=0.143). In another 6 in each group, eradication was achieved with rescue treatment. Recurrence of Pa occurred mainly in the first months following drug treatment.

Conclusion: Both regimens tested have a high and comparable immediate Pa eradication rate. Only half of the patients remain free of Pa during 6 months without additional antibiotic treatment. Pseudomonas aeruginosa hemolytic phospholipase C (PlcH) is an extracellular virulence factor that cleaves the phosphorylcholine head group from phosphatidylcholine and sphingomyelin, both present in pulmonary surfactant and lung cell plasma membranes. In vitro and in vivo experiments have demonstrated PlcH alteration of lung surfactant, immune cell function, and inflammatory processes. In cystic fibrosis (CF)-related P. aeruginosa infections, early and robust antibody production to PlcH indicates that this secreted virulence factor is produced in the CF lung. We previously identified the P. aeruginosa transcription factor GbdR as the primary regulator of plcH transcription in the mouse lung. Transcriptional induction by GbdR is dependent on glycine betaine (GB) concentration within the bacterial cell. The enzymes required for GB catabolism to dimethylglycine in P. aeruginosa are GbcA and GbcB. To test the hypothesis that plcH transcription and resulting phospholipase activity could be reduced by perturbing intracellular levels of GB, we generated gbcAB deletion strains and inducible gbcAB expression strains. Here we show that changes in the rate of GbcAB-dependent catabolism of GB alter both plcH transcription and resulting PlcH-dependent virulence as measured by transcriptional fusions, enzymatic assay, epithelial cell biofilm formation, and mouse lung infection data. Our data also support the hypothesis that the native mechanism controlling the size of the cytoplasmic GB pool in P. aeruginosa is based on regulation of gbcB translation. By understanding the connection between catabolism of host-acquired molecules and the regulation of virulence factors we gain insight that will be valuable in developing novel therapeutic agents that target microbial catabolic pathways in order to reduce bacterial virulence and improve patient lung function. P. aeruginosa-induced activation of NF-κB and secretion of proinflammatory cytokines by airway epithelia requires that the bacteria express flagellin. We tested whether P. aeruginosa and human airway epithelia secreted factors that modulated this response. Experiments were performed on both the Calu-3 cell line and primary cultures of tracheal epithelia. P. aeruginosa strain PAKfliC (flagellin knock-out) did not activate NF-κB or IL8, but inhibited flagellin-activated NF-κB by 40-50% and IL8 secretion by 20-25%. PAKfliC also inhibited NF-κB induced by IL1β and toll-like receptor 2 agonist Pam3CSK4. Similar inhibitions were observed with strains PAK, PAO1 and PA14. The inhibitory factor was present in conditioned medium isolated from PAKfliC or Calu-3+PAKfliC, but it was not present in conditioned medium isolated from Calu-3 cells alone or from PAKfliC that had been heat-treated. Inhibition by PAKfliC-conditioned medium was exerted from either apical or basolateral side of the epithelium; was enhanced in simple Ringer's solution compared to in tissue culture media; and did not result from altered pH or depletion of glucose. The inhibitory effect of conditioned medium was abolished by boiling and appeared from filtration studies to result from effects of a factor with molecular weight <3 kDa. These and further studies with isogenic mutants lead to the conclusion that the NF-κB and IL8 response of airway epithelia to P. aeruginosa results from a balance of proinflammatory effects of flagellin and anti-inflammatory effects of a small (<3kDa), heat-sensitive factor(s) that is not LPS, C12 homoserine lactone, alginate, CIF or exotoxins A, S, T, U, or Y. P. aeruginosa infections of lung airways are characteristic of cystic fibrosis (CF). Flagellin and quorum-sensing homoserine lactones (N-(3oxo-dodecanoyl)-S-homoserine lactone, 3O-C12, and N-butyryl homoserine lactone, C4) are released from P. aeruginosa into the airway surface liquid of infected individuals. We tested flagellin (10 -6 g/ml) and 3O-C12 and C4 (1 to 100 µM) on Clsecretion and Ca 2+ and cyclic-AMP responses of Calu-3 cells, a human, non-CF, serous-like line. Flagellin and 3O-C12 (but not C4) stimulated slow increases in Clsecretion (I Cl , Ussing chamber transepithelial electrophysiology) that began within 1-5 mins and continued to increase to 30-150 µA/cm 2 over 45 mins. Flagellin and 3O-C12 activations of I Cl were: largely blocked by CFTR inhibitor CFTRinh172, somewhat larger for 3O-C12 than for flagellin and smaller than those elicited by forskolin. Stimulatory effects of flagellin on I Cl were abolished by prior stimulation with 3O-C12, indicating common mechanisms of action. 3O-C12, but not C4 or flagellin, increased cytosolic [Ca 2+ ] (Ca i , fura-2 fluorescence imaging in single cells), and there was similar concentration dependence for effects on Ca i and I Cl . Thapsigargin, an inhibitor of CaATPase that releases Ca 2+ stored in the endoplasmic reticulum, increased both Ca i and I Cl . Subsequent stimulation with 3O-C12 elicited no further stimulation of Ca i but small, further increase in I Cl . 3O-C12 also increased Ca i and I Cl , and prior treatment of cells with 3O-C12 prevented further stimulatory effects of thapsigargin on I Cl and Ca i . 3O-C12 and flagellin also increased cytosolic [cAMP] as measured by a sensitive FRET-based cAMP sensor in single cells; stimulatory effects of flagellin were smaller than those of 3O-C12, consistent with effects of flagellin and 3O-C12 on I Cl . I Cl stimulated by 3O-C12 or forskolin was inhibited by about 50% by prior treatment with the cAMP antagonist RpBrcAMPS and increased by >50% by treatment with the phosphodiesterase inhibitor isobutylmethylxanthine (100 µM). These results indicate that, in addition to having proinflammatory effects to activate NF−κB signaling and IL8 secretion in airway epithelia, flagellin and 3O-C12 both stimulate Clsecretion that is CFTR-dependent and is mediated at least partly through agonist-triggered increases in cytosolic [cAMP]. 3O-C12, but not flagellin, may also stimulate Clsecretion through its effects to raise Ca i . In CF, flagellin and 3O-C12-triggered signaling and cytokine responses will be retained but activation of CFTR-dependent Cland fluid secretion will be prevented. Support: CFF and CFRI. Pseudomonas aeruginosa has enormous clinical relevance to patients with cystic fibrosis (CF), many of whose lungs are colonized by this bacterium in the first few years of life. The strains involved are generally consid-ered environmental, and yet to date the environmental niches that are most important in lung colonization remain unknown. There are two possible benefits to identifying these niches. First, if P. aeruginosa or the genotypes most likely to colonize the lung occur in a limited set of habitats, infection risk could be reduced and colonization delayed by simple interventions (cleaning, avoiding specific exposures). Second, identification of infecting niches may reveal characteristics of good colonizers that could be targeted with new therapeutics for early infections.

In this study we examine the distribution of bacteria in the genus Pseudomonas among types of niches from which the bacterium is highly likely to have access to the lungs of a young child with CF: the human home. We sampled up to 100 niches in each of 30 homes of healthy adult non-CF volunteers, about 26% of whom had children at home. These niches include both human and pet niches (i.e. nose, mouth, eyes, etc), as well as kitchen, bathroom, children's play areas, etc; in all, 1050 samples were taken with swabs and streaked onto Pseudomonas Isolation Agar. Where growth occurred, the isolate was identified by its 16s rDNA sequence. In all, we isolated 146 Pseudomonas, 22 of which were P. aeruginosa. We found that Pseudomonas putida was the most commonly isolated Pseudomonas and was typically found in house plants and soils. Pseudomonas aeruginosa was most frequently isolated from drains (12/22 isolates). Both P. aeruginosa and P. putida were isolated from healthy humans (fecal, skin, navel, and scalp), and P. putida was also isolated from pets. Other Pseudomonas species were also isolated from healthy humans (mouth, throat, and from underneath of fingernails). We investigated variability in the distributions of genotypes and phenotypes within species using multi-locus sequence typing and phenotypic assays from API strips (bioMéreux), comparing isolates both within and between different homes. We discuss how these results have influenced the design of an ongoing sampling study involving comparisons of clinical isolates from children with CF and isolates from the households in which they live.

Ledson, M.J.; Dyce, P.; Cowperthwaite, C.; Spencer, E.; Walshaw, M.J.

Many cystic fibrosis (CF) patients in the UK became chronically infected with the highly transmissible and often fatal pulmonary pathogen, Burkholderia cenocepacia, in the early part of the 1990s, probably following its transfer into the UK CF community from visitors to North American Summer Camps in the late 1980s. However, effective segregation coupled with better microbiological surveillance has prevented further spread of this epidemic. When our adult unit opened in 1993, we accepted a cohort of soinfected patients mainly from the paediatric sector. Four further patients attending our large adult unit (now 250 patients) have become infected with Burkholderia cenocepacia since then, in three cases due to social contact outside the hospital environment and in the remaining case there was no evidence to suggest cross infection despite extensive investigation. We now report the outcome for all our Burkholderia cenocepacia patients, 16 years later. [2 to 180] , 6 male), there has been little alteration in clinical state over the last 6 years (mean FEV1 change -0.75% per year), suggesting they are a "survivor" population. This group includes one CF person who visited the original summer camps in 1988 (who therefore may have been an index case), and whose clinical state remains unchanged, 21 years later. These data indicate that the epidemic of Burkholderia cenocepacia in Merseyside is coming to an end, through the inevitable increased mortality caused by this organism coupled with the prevention of new cases by effective strict segregation policies in both adult and paediatric sectors.

Perrin, F.M. 1, 3 ; Prendergast, C. 2 Background: Nontuberculous mycobacteria (NTM) are seen increasingly as pulmonary pathogens in patients with cystic fibrosis (CF), but their significance in sputum is unclear as they may represent airway colonisation or disease. Anecdotally Mycobacterium abscessus (M. abscessus) is associated with more fulminant disease and less good outcome than other NTM species. Indicators highlighting which patients are likely to develop disease would be invaluable. The morphological appearance of M. abscessus colonies on culture, reported as rough or smooth, may contribute to this. It has been speculated that a rough appearance may indicate more invasive disease with individual case reports describing transformation of cultures from a smooth to rough morphotype in patients with progressive deterioration. The aim of this study was to compare clinical state with colony morphology in patients with M. abscessus and CF.

Method: All patients seen at the Royal Brompton Hospital Adult CF Centre, culturing M. abscessus in their sputum in 2008, were identified from the centre database. From September 2008 all sputum culturing M. abscessus was grown on blood agar and colony morphology noted by a single observer. Patients who met ATS microbiological criteria for NTM (positive cultures on 2 separate occasions) were classified as having "disease" if their physician had initiated treatment or "colonisation" if not. This was correlated with colony appearance.

Results: M. abscessus was isolated from 27 patients. In 7 patients this was on one isolate only (or 2 isolates less than 2 weeks apart); the colony morphology was studied in 3 of these and all were smooth in appearance. Subsequent samples from these patients were culture negative. The remaining 20 patients met ATS microbiological criteria, with M. abscessus isolated on 2 or more occasions. The colony morphology of 15 was examined (1 was excluded due to an equivocal result). In 10 patients 2 or more consecutive samples were studied. Eight patients (57%) were on treatment. The table correlates colony appearance and clinical state for these patients meeting ATS criteria.

Conclusion: Patients with only one positive culture had smooth colonies, suggesting this morphotype may represent transient isolates. Patients who met ATS microbiological criteria predominantly had rough colonies; disease could not be distinguished from colonisation in these patients by colony morphology. Moreover, the presence of smooth colonies, and conversion of colonies from rough to smooth, in patients on treatment is unexpected if a rough appearance is truly associated with more severe disease. Further prospective follow-up studies are required to delineate the relationship between M. abscessus colony morphology, virulence and strain type as well as the role of colony type on biofilm formation. 1 and SCr, monitor patient demographics, individual pharmacokinetic parameters, and outcome measures associated with adequate resolution of bacterial infection during high-dose twice-daily tobramcyin therapy in an attempt to gain an increased understanding of the most optimal way to administer aminoglycosides in an acute cystic fibrosis exacerbation; in particular, during twice-daily dosing schedule routinely utilized at the University of Kentucky since 2005.

Methods: Medical records of all patients ≤ 18 years of age who were admitted to the University of Kentucky between January 1, 2007 and December 31, 2008 and received twice daily intravenous tobramycin therapy for treatment of an acute cystic fibrosis exacerbation were reviewed.

Patients' demographics, pharmacokinetic parameters, pulmonary function tests, laboratory values, microbiological data, in-hospital course at baseline and after therapy and incidence of adverse effects were documented.

Results: Of the 39 patients reviewed 21 were male, 18 were female between the ages of 6 to 18 years. Data analysis revealed an average height and weight of 145 cm and 37.75 kg respectively. The average tobramycin dose was 12.42 ± 2.7 mg/kg/day which resulted in a mean four and ten hour levels of 3.7 mcg/mL and 0.7 mcg/mL. The average FEV 1 during tobramycin therapy was 58.1‰. Mean FEV 1 (‰) treatment effect per year regardless of disease severity was a net gain in FEV 1 ranging between 1.93 to 6.87‰ over a median length of follow-up of 8.7 months. The average SCr during therapy for all age groups was 0.56 mg/dL. The mean change in SCr per year and over the entire treatment course was an overall decline by 0.004 and 0.018 mg/dL respectively. Individual pharmacokinetic parameters will also be assessed.

Conclusions: Average treatment effect for all age groups showed an average increase in FEV 1 per year ranging between 1.93 to 6.87‰. High dose extended interval twice-daily tobramycin therapy does not appear to produce significant increase in SCr, with an average decline in SCr per year and over entire study treatment time period. Data analysis of individual pharmacokinetic parameters is ongoing. The results of this study show that high-dose twice-daily tobramycin therapy is a plausible treatment option for an acute cystic fibrosis exacerbation in pediatric patients with normal renal function ≤ 18 years of age. . Traditional antibiotics are losing efficacy against many prevalent organisms, including Pseudomonas aeruginosa (Pa), highlighting the need for alternative therapeutic approaches. The endogenous antimicrobial biocatalyst lysozyme (LYZ) has antimicrobial activity against gram-positive and gramnegative pathogens. However, its highly cationic properties and electrostatic interactions with airway inflammatory polyanions (double-stranded DNA, F-actin, and mucin) limit its activity in the context of the infected lung. We hypothesized that remodeling the electrostatic surface potential of LYZ could produce an antimicrobial protein capable of evading electrostatic aggregation while maintaining high levels of lytic activity, and that such an enzyme would promote more rapid clearance of Pa from the lung.

Methods: First, eight poorly-conserved, surface-exposed, positively charged amino acids on LYZ were combinatorially mutated to alternative amino acids of either neutral or negative charge. Using novel high-throughput screening, variants that retained catalytic antimicrobial activity in the presence of polyanions were identified. Next, the in vivo antimicrobial efficacies of the engineered and wild type LYZ were compared by administration of LYZ one hour after C57Bl/6 lung infection with mucoid Pa (FRD-1).

Results: Over 100 promising engineered LYZ variants that retain their antimicrobial efficacy in the presence of polyanions have been developed to date. "2-3-7" was selected for in vivo testing based on its in vitro activity against Pa. The administration of "2-3-7", when compared to wild type LYZ, to an infected mouse lung results in a dose-dependent reduction of Pa CFUs despite the presence of high levels of airway polyanions. Background: Published prevalence rates of filamentous fungi recovery from airway cultures in individuals with cystic fibrosis (CF) range from 16-57%. We have previously shown that isolation of filamentous fungi is associated with a lower absolute FEV1%, but the effect on lung function decline is unknown.

Methods: We conducted a retrospective cohort study of all individuals seen at the Johns Hopkins (JH) CF center from 1997-2007, who had microbiologic and clinical data records. Children were defined as individuals 6-18 years. Fungal colonization was classified as periods when the same fungal species was cultured during two or more quarters occurring no more than 1 quarter apart. We developed two multiple linear regression (MLR) models, utilizing generalized estimating equations (GEE) to investigate associations between filamentous fungi and lung function decline (FEV1% per year) with adjustment for confounders.

Results: A total of 448 individuals 6+ years old were enrolled in the study 1997-2007. First, in 251 children, lung function decline was found to be more rapid during quarters with isolation of filamentous fungi compared to quarters without, even after adjusting for confounders (difference FEV1% per year -0.41; 95% CI -0.82 to -0.01; p=0.046). No significant difference was seen in 255 adults (difference FEV1% per year 0.10; 95% CI -0.004 to 0.20; p= 0.060).

Next, we investigated the dynamics of lung function before, during, and after fungal colonization in children stratified by ABPA diagnosis. For 31 fungal-colonized children without ABPA diagnosis; the average decline in lung function before colonization was -3.58 FEV1% per year and there was no significant difference in decline during fungal colonization (difference FEV1% per year -0.23; 95% CI -1.33 to 0.86; p=0.676) or post-colonization (difference FEV1% per year -0.28; 95% CI -1.60 to 1.04; p=0.676). Fungalcolonized children without ABPA had significantly lower average FEV1% (79.68 vs. 87.81; p<.001) and greater number of clinic visits per quarter (2.13 vs. 1.52; p<.001) compared to children never colonized with filamentous fungi.

Fungal-colonized children diagnosed with ABPA (n=5) had an average decline in lung function before colonization onset of -1.94 FEV1% per year and during colonization lung function declined at a significantly greater rate (difference FEV1% per year -1.67; 95% CI -3.27 to -0.07; p=0.041), adjusted for confounders. The post-colonization period was also associated with a significantly greater decline in lung function as compared to before colonization (difference FEV1% per year -2.18; 95% CI -3.76 to -0.60; p=0.007).

Conclusions: Fungal colonization, in CF children without ABPA does not appear to be associated with more rapid decline in lung function, but acts as a marker of disease severity. In children with ABPA, even after adjustment for treatment, lung function declined at significantly more rapid rate during and post fungal colonization, as has been previously reported. Future studies should look at the impact of individual fungal species on lung function and ABPA treatment regimens that may reverse lung function decline. UZ Brussel, Brussels, Belgium; 2. Microbiology, UZ Brussel, Brussels, Belgium Background: In cystic fibrosis (CF) care, bacterial pathogens known to colonise the respiratory tract are usually thought to be also responsible for acute pulmonary exacerbations. Viral involvement is poorly evaluated, possibly due to the lack of sensitive detection methods. As respiratory viruses are known to replicate mainly in the nasopharyngeal cavity, nasopharyngeal secretions are considered preferential samples for viral detection.

Aims: To evaluate the involvement of viruses in CF pulmonary exacerbations and to identify the main viral pathogens.

Methods: At the onset of any acute pulmonary exacerbation we prospectively collected respiratory secretions obtained by nasal wash, to detect respiratory viruses using multiplex-PCR and viral culture. Blood for viral serology was also collected if possible. Whenever flexible fibroscopy with bronchoalveolar lavage (BAL) was indicated, BAL-fluid was also collected for detection of viral pathogens by multiplex-PCR and culture.

Results: For 199 acute pulmonary exacerbations in 86 CF patients (median age = 11y 4m, sex ratio: F/M = 42/44), a viral pathogen was detected in 64/199 (32.2%) episodes and in 3/64 episodes 2 viruses were found. The 3 main viral pathogens found were influenza A virus, respiratory syncitial virus and human metapneumovirus, respectively in 15, 15 and 8 episodes.

In 56/64 (87.5%) episodes the viral pathogen was detected on respiratory secretions, in 52/56 (92.8%) cases on nasal washes and in 4/7 (57.1%) cases on BAL-fluid. Viral serology was only positive in 9/30 (30%) episodes.

Conclusions: Common respiratory viruses seem to play a role in an unexpected high proportion (32%) of pulmonary exacerbations in CFpatients of all age groups.

PCR and culture on respiratory secretions obtained by nasal wash is a rapid, sensitive and non-invasive method for viral detection compared to serology. Background: Annual influenza vaccination (InV) is recommended for risk-groups, including cystic fibrosis (CF) patients. Although the immunogenic effect of InV in CF is thought to be comparable to that in healthy individuals, InV efficacy in CF is not well documented. We reported 2 case reports of InV failure in CF patients in 2005 (J Cyst Fibros 2005;4 suppl).

Aims: To evaluate the impact of influenza infection in well-vaccinated CF patients, using commercial trivalent inactivated influenza vaccine (TIV), European formulation.

Methods: During 4 winter seasons, we prospectively collected nasal secretions by nasal wash for PCR and culture and/or blood for serology, for detection of respiratory viruses in all CF patients that presented with an acute pulmonary exacerbation. All CF patients from the age of 6 months were vaccinated against influenza, according to ACIP recommendations.

Results: Influenza viruses were detected in a total of 21/170 (12.4%) acute pulmonary exacerbations in 75 CF patients (median age = 17y7m; sex ratio: F/M = 38/37). Influenza infections were mainly found in adolescent and adult patients, with a median age of 20y 7m (age range: 11m-37y1m) in the infected group. Influenza A was found in 19 and influenza B in 2 episodes. In 13/19 influenza A and 1 influenza B infections the patient was febrile (body temperature >38.5∞C) and in 8/19 influenza A infections the patient needed (prolonged) oxygen supplements. For 17/21 influenza infections hospitalisation was necessary with a mean stay of 11.7 days. Treatment with Oseltamivir was given if diagnosis was confirmed within 5 days of onset, and was well tolerated.

Conclusions: Influenza infection is frequent in CF exacerbations, especially in adolescents and adults, despite correct vaccination. Influenza in CF patients is associated with a high rate of oxygen need and prolonged hospi-talisation. Subclinical cases might be missed in this study. More data are needed to evaluate the effect of Oseltamivir on the bad course.

Malek, A.A.; Hogan, D. Department of Microbiology and Immunology, Dartmouth Medical School, Hanover, NH, USA Phosphatidylcholine (PC) is abundant in the lung environment as a major phospholipid component of surfactant fluid and cell membranes. Pseudomonas aeruginosa utilizes two enzymes: phospholipase C (PlcH) and phosphorylcholine phosphatase (PchP), which help break down PC to yield diacylglycerol and choline. Work from our lab and others has shown that the liberated choline promotes P. aeruginosa virulence and survival in a variety of ways. Despite the multifaceted roles of choline, the mechanisms for its uptake by P. aeruginosa in the lung are not yet understood. Using whole genome analysis and homology based searches in combination with mutant analyses, we identified one predicted ABC transporter, Bcc, and one putative choline symporter, BetT1, that together appear to mediate uptake of choline under physiological conditions. Additionally, our findings suggest that the transcriptional regulation of these genes is different. On availability of the choline catabolite, a regulator GbdR highly induces the expression of bcc genes. In contrast, the expression of betT1 gene does not depend on GbdR. Current work is focusing on the role of these transporters in uptake of choline for the purpose of surface modification and the regulation of choline induced virulence genes in vivo. We are testing the latter by analyzing the expression the of plcH, pchP, betA and betB genes (genes highly expressed by P. aeruginosa in the CF lung) as molecular markers of choline uptake in lung surfactant fluid and mouse model of infection. A thorough understanding of the choline transporters will facilitate design of novel anti-Pseudomonas therapies targeted at decreasing virulence in the cystic fibrosis lung. In 1995 over 75% of the children attending the Liverpool Regional CF Centre had Pseudomonas aeruginosa (PA) isolated from their respiratory culture. In the majority of these children (n, 55), this was identified as a genetically identical clone, assumed to be a transmissible strain (supported by the subsequent recognition of the Liverpool Epidemic Strain (LES) in numerous CF centres in the UK and beyond). Initially LES was considered inherently resistant to ceftazidime (a feature that prompted the initial investigation and genotyping), however sensitive isolates have subsequently been recognized. Evidence suggests that LES may be associated with an accelerated decline in clinical condition. The aim of this project was to determine the impact of strict inpatient and outpatient segregation on the prevalence and phenotype of this strain in the Liverpool Paediatric clinic.

Method: A retrospective case note review. Since 2003, a PCR assay has been used to confirm rapid identification of the LES and this has informed the segregation policy. For information on the PCR assay, please contact Dr Craig Winstanley (c.winstanley@liv.ac.uk).

Results: In 2003, 47 children in the clinic had LES isolated from their respiratory cultures. In 2009, 9 children have LES regularly isolated. Since 2003 two patients have acquired LES, in one this was first grown at bronchoscopy and has been consistently identified since and in the other, LES was identified on the first culture following a late diagnosis. Two patients who grew LES consistently, now grow a non-LES strain. Of the other patients two died in our care, one received a lung transplant and the rest were transferred to the adult clinic at 17 years of age. In 35-45% of the chil-dren infected with LES from 2003-2009, the strain was sensitive to ceftazidime (colistin resistance ranged from 0% to 3%). In children with non-LES PA there was virtually no ceftazidime resistance (0% to 1.6%) identified in our clinic population.

Conclusion: A strict segregation policy has significantly reduced the prevalence of LES in the Paediatric Clinic in which it was first recognised. This policy has been aided by the use of a rapid PCR assay. The previous segregation policy (based on ceftazidime resistance) is not completely reliable. Although most LES are ceftazidime resistant, this should not be considered an inherent property of the strain. A low threshold of suspicion with respect to the LES should be exercised when considering your clinic population. Factors that should prompt further investigation include a high prevalence of PA infection and unusual patterns of resistance. Background: Literature is emerging describing zoonotic transmission of bacteria and viruses to humans. Bordetella bronchiseptica (BBS) is a common cause of "kennel cough" in dogs. Despite a large proportion of people keeping dogs as pets BBS does not appear to be responsible for significant respiratory morbidity except in compromised health conditions. This is in contrast to its descendants Bordetella pertussis and Bordetella parapertussis.

In cystic fibrosis (CF) increased microbial surveillance has led to newer organisms being identified within the lung. Of note BBS has been isolated as a commensal in the sputa of CF patients. The clinical implication of infection by this organism remains unknown.

Aims: To review the frequency of BBS in our CF clinic and report the clinical features of patients in whom BBS was isolated.

Methods: In depth chart review of clinical features, clinical course, copathogens and lung function of patients with BBS.

Results: Since 2005, BBS was isolated in 3 (2.5%) of 120 clinic patients at BCCH CF Clinic. All patients had contact with sick dogs. The clinical features and change from baseline lung function at presentation and following BBS isolation are detailed in table 1. Symptoms varied from asymptomatic to an acute respiratory exacerbation requiring admission for intravenous antibiotic therapy. A year following infection lung function was reduced in both patients 2 and 3. This improved in patient 2 who remains colonized with BBS 4 years after isolation (2009 Lung function: FVC 115%, FEV 1 105%). Patient 3 was transferred to the adult service and re-isolated BBS 23 months following initial culture (2009 Lung function: FVC 83%, FEV 1 53%). Summary: We have isolated Bordetella bronchiseptica in a total of three patients in our clinic. The infection is highly likely to have occurred from contact with dogs infected with BBS. Clinical features may vary but infection resulted in one patient chronically colonized with Bordetella bronchiseptica. The impact on lung health remains to be determined. Background: Non-tuberculous mycobacteria (NTM) are emerging as increasingly common isolates in the sputum of cystic fibrosis (CF) patients, and some suffer significant clinical deterioration following NTM infection. Currently there exists little guidance or consensus on how and when to treat NTM infection.

Aims: This study aimed to establish the scale of the problem of NTM infection in the UK CF community, and how proactively it is screened for and managed. We wished to a) estimate the 2 year prevalence of NTM isolation b) determine the frequency of screening for NTM and c) delineate how NTM isolation is subsequently managed.

Methods: A one-page questionnaire was sent to clinicians at all recognised (23 adult, 29 paediatric) CF centres throughout the UK. Questions were asked on numbers of CF patients, numbers with NTM and management of these patients.

Results: Responses were received from 17 and 22 of these centres respectively. These constituted a total of 3230 adult, and 3207 paediatric CF patients. Within this sample 169 adults and 95 children (5.2% and 4.4% respectively) had cultured NTM in the past 2 years, of which 118 adults and 71 children (3.7% and 3.2% respectively) had cultured NTM on at least 2 separate occasions. The majority of responding centres (13 adult, 16 paediatric) cultured for NTM only yearly, with most of the remaining centres testing for NTM only if there was a specific indication eg. clinical deterioration. The NTM pathogens most frequently cultured in the paediatric and adult groups were the rapidly growing Mycobacterium chelonae/Mycobacterium abscessus (56% of adults and 80% of children culturing NTM) followed by Mycobacterium avium complex (MAC) (31% of both adults and children). Other NTM less commonly found included M. gordonae, M. xenopi, M. fortuitum, M. mucogenicum and M. kansasii. Criteria for starting treatment varied widely between centres, with some ready to commence on culture +/smear positivity only, while others required clear clinical deterioration and/or radiological changes. Overall, 40% of adult and 55% of children who had cultured NTM over the past 2 years had been commenced on specific.antibiotic treatment. Six adult CF patients and 2 paediatric CF patients had been refused a lung transplant on the basis of persistent NTM infection.

Discussion: Our findings differ from those reported in the US by Olivier (Am J Respir Crit Care Med 2003;167:828) in showing an apparently lower prevalence (5% vs 13%) and a much higher proportion of rapid-growing mycobacteria rather than MAC. The apparent lower prevalence may, at least in part, be due to a low frequency of testing for NTM. About half of those culturing NTM had been started on treatment, though the threshold for commencing treatment varied across centres. The diagnosis of NTM infection has a significant impact for CF patients, both in the requirement for long-term therapies and potential refusal for transplant. Further research to guide therapeutic strategies is urgently needed. Objective: Ralstonia species have recently been isolated from respiratory tract of cystic fibrosis (CF) patients and most were identified as Ralstonia mannitolilytica (RM). These gram-negative bacilli can be misidentified as Burkholderia cepacia (BC) complex. The clinical impact of infection with RM in CF patients is unknown. We conducted a retrospective analysis of the clinical evolution of CF patients who acquired RM in their airway secretions.

Methods: We identified CF patients who had positive culture for RM from January 2007 to April 2009 using data provided by the microbiology laboratory. We carried out a chart review of these patients to collect information that would help to describe their clinical evolution.

Results: Of the 301 patients followed in our CF clinic, six (2 males and 4 females) had positive RM culture. Their age ranged from 19 to 38 years and 3 out of 6 patients were homozygous for ∆F508. The other three patients were heterozygotes (∆F508/I148T, ∆F508/621+1GÇT and 621+1GÇT/G542X). At the time of their first positive RM culture, their mean FEV1 was 49% predicted (range 42-58%) and their median body mass index was 19.2 kg/m 2 (range 18. 5-23.2) . The year preceding the acquisition of RM, the median time spent in hospital was 43 days (range 22-159). Prior to the acquisition of RM, one patient was colonized with Ralstonia Picketti (formerly called Burkholderia picketti), two with multiresistant Pseudomonas aeruginosa, one with BC and three with methicillin-resistant Staphylococcus aureus. After reviewing the social contacts and the clinic visits of these patients, we do not believe that there was any person-to-person transmission of RM. Within two months of the first positive culture for RM, four patients presented episodes of fever up to 39∞C. Three patients developed marked leucocytosis (white cell count ≥ 40x10 9 /L). For three patients, RM was identified in other specimens than sputum (bacteremia (1), sinus (1) and mediastinal lymph node biopsy (1)). Because it was a multiresistant bacteria, prolonged course of combination intravenous antibiotic therapy was necessary to control the infection in most patients. Despite treatment, three patients died. Two of them died within 3 months of the acquisition of RM. Both presented with acute pulmonary exacerbation that rapidly evolved to hypoxemic respiratory failure for one patient and sepsis complicated by multiple organ failure (renal and lung) for the other. The third patient died of progressive hypercapnic respiratory failure. Eradication of RM occurred only for one of the three remaining patients, after one month of antibiotic therapy.

Conclusion: Although infection with Ralstonia mannitolilytica is rare (6/301patients), our case review suggests that acquisition of this bacteria might accelerate the course of CF disease as it has been documented with some of the genomovars of the Burkholderia cepacia complex. Further studies in larger cohorts will be necessary to characterize the impact of the acquisition of Ralstonia mannitolilytica on CF disease. 

Objectives: The development of antibiotic resistance of Pseudomonas aeruginosa (PA) is a significant clinical concern in cystic fibrosis (CF) patients, particularly in the expanding adult population. Antibiotic efflux pumps are a known mechanism by which PA acquires antibiotic resistance and has been suggested to play a major role in PA infections in patients with CF. Resistance to aminoglycosides (AG) has specifically been associated with increased expression of the mexXY-OprM efflux pump in both CF and non-CF isolates. However, little is known about the time course and factors that facilitate increased mexXY expression and whether it is associated with deterioration in lung function. We hypothesize that mexXY mRNA expression would increase with longer duration of PA infection in CF patients and that it would be associated with worse lung function.

Methods: Serial clinical isolates of PA were obtained over 3 years from both pediatric and adult CF patients for evaluation. Five PA isolates were obtained from each sputum sample and stored for analysis. Antibiotic sensitivities were obtained by conventional clinical microbiology testing for each sputum sample using CF Foundation guidelines. Subjects were divided into 3 patient cohorts including adults (> 18y) and children with either 1st positive PA culture or those chronically infected (> 2 sputum PA (+) >1 year duration). PA isolates were grown to log phase and mRNA was obtained. Quantitative real-time PCR was used to determine mexXY mRNA expression in clinical isolates compared to wild-type strain PAO1. Previous studies have shown that mexXY mRNA expression correlates well with overall mexXY protein expression. We also have data on duration of PA infection and longitudinal change in lung function and detailed antibiotic exposure histories for each patient which will be compared to mexXY mRNA expression.

Results: We have obtained a collection of approximately 2000 clinical PA isolates from CF patients at various stages of infection and longitudinally within the same patient. We have evaluated a select group of clinical PA isolates for mexXY mRNA expression to date. We have noted an approximate 5-fold increase in level of mexX mRNA expression in isolates from AG-resistant sputum samples compared to isolates from AG-sensitive samples when compared to wild-type strain PAO1 (p<0.05). When samples were subdivided into patient cohorts, mexXY mRNA expression tended to be higher in patients with longer duration of PA infection.

Conclusions: Our preliminary data suggests that mexXY mRNA expression may play a role in clinical AG resistance in CF PA infection. Duration of PA infection may be associated with increased levels of mexXY expression in the CF host. Background: Pseudomonas aeruginosa (PA) is an important pathogen in cystic fibrosis (CF). The type III protein secretion system is a virulence factor of PA that is important in hospital-acquired infections but appears to be selected against in CF. Previous reports indicated that PA strains transiently infect the airways of CF patients prior to a single strain establishing a chronic infection. The factors that determine which strain will successfully persist rather than be cleared are poorly understood. We examined whether type III secretion phenotypes affected the ability of PA strains to persist during initial infection of CF patients.

Methods: Twenty-six pediatric patients (0-16 years of age; mean 5.5 years. 11 boys, 15 girls) in whom PA was isolated for the first time after at least 2 years of negative cultures were considered newly infected and were included in the study. Five PA isolates were characterized from each respiratory culture. The type III secretion phenotypes of the isolates infecting these patients were determined by immunoblot analysis of culture supernatants grown under secretion-inducing conditions and type III secretion genotype by multiplex PCR. PA strain genotypes were determined by Random amplified Polymorphic DNA typing (RAPD). Fisher's exact test was used for all comparisons.

Results: All PA strains contained at least one of the genes encoding type III secreted proteins, although only 9 of 31 (29%) of the isolates secreted detectable amounts of these proteins. In agreement with previous studies of CF patients, most strains lacked the exoU gene but contained the exoT and/or exoS genes.

Eight of 13 (61.5%) type III secretion negative strains persisted vs. 3 of 12 (25%) type III secretion positive strains. In six patients, phenotypically positive and negative strains co-existed in the same respiratory sample, and in 5 of these 6 cases (83.3%), the strain persisted. These differences were not statistically significant.

Conclusions: The type III secretion profile of isolates was not associated with persistence of PA in this small cohort of newly infected CF patients. Larger studies are needed to elucidate more subtle effects of type III secretion on PA persistence in CF. Objective: We sought to determine whether initiation of an aggressive antibiotic eradication regimen upon discovery of mucoid PsA in the CF airway can be successful in eradicating mucoid PsA from the CF lung. The primary goal of this study was to determine whether newly cultured mucoid PsA can be eradicated from the CF airway and to determine the success rate of the typical therapeutic regimen at our institution.

Methods: We performed a retrospective analysis of all patients with CF, cared for at our institution, with growth of mucoid PsA in airway culture and who underwent an eradication attempt between January 2003 and December 2008. Data regarding age of CF diagnosis, age at mucoid PsA isolation, forced expiratory volume in 1 second (FEV 1 ), body-mass index (BMI), gender, therapeutic interventions, and microbiologic data were obtained from the charts. Eradication of mucoid PsA was the primary endpoint and was defined as 3 airway cultures without demonstrable growth of mucoid PsA within 6 months of the eradication attempt and while off antibiotics. Data were evaluated using student's t-test or Mann-Whitney U test, as applicable. Kaplan-Meier survival analysis was performed to discern time-dependence of PSA culture negative survival. Linear regression was performed with lung function and BMI data.

Results: Of the 355 patients with CF, 241 had at least one PsA positive airway culture, 128 had a new culture with mucoid PsA, and 14 of these underwent an eradication attempt. Eradication of both mucoid and nonmucoid PsA was observed in 6 of 14 patients and elimination of only mucoid PsA in an additional 4 patients. Prolonged (>24 months) PsA-free cultures were seen in 21.4% of patients, whereas 64.3% of patients remained free of mucoid (but not non-mucoid) PsA. Patients who did not undergo an eradication attempt had significantly lower percent-predicted FEV 1 at the study conclusion when compared to the patients who underwent an eradication attempt (68.1 ± 2.6 vs. 90.8 ± 6.0, p<0.005). Additionally, the patients who did not undergo an eradication attempt had a more pronounced decline in lung function based on linear regression (slope -0.038 vs. 0.462, p=0.04).

Conclusions: Elimination of mucoid PsA from the airways of select CF patients is possible with current antibiotic regimens. This strategy may be considered in appropriately selected CF patients who demonstrate new growth of mucoid PsA. 

Background: In children with therapy-refractory bronchitis or pneumonia, bronchial Chlamydophila (C.) pneumoniae infection is common and associated with pathological lung function. Cystic fibrosis (CF) is a hereditary illness and chronic respiratory tract symptoms as well. The aim was to study the prevalence of C. pneumoniae infection and its potential association with progression in CF.

In a prospective multicenter study, C. pneumoniae was detected in sputum by polymerase chain reaction with enzyme immunoassay detection and in paired serum samples with microimmunofluorescence test. Pseudomonas (P.) aeruginosa was cultured in sputum. Lung function tests and clinical score characterized impairment of CF.

Results: C. pneumoniae infection was detected in 22 of 63 CF patients. It was strongly associated with P. aeruginosa co-infection (21/22 versus 23/41; p=0.001) and obstructive disturbance in 60 cooperative patients with pulmonary function tests (20/20 versus 29/40; p=0.01). In 22 patients with at least six months follow-up a trend line was calculated for every lung function parameter and clinical score by linear logistic regression including all tests. Six months loss of FEV1 was higher in sixteen C. pneumoniae positives (-2.11 [-5.04 Conclusion: C. pneumoniae infection is common in CF. In P. aeruginosa positives, a C. pneumoniae co-infection is associated with a faster progression in CF. 

Methicillin-resistant Staphylococcus aureus (MRSA) is now a worldwide public health problem due to the increasing rate of infection in several settings as well as in cystic fibrosis (CF) patients. MRSA was first recognized as being acquired from hospitalized patients (HA-MRSA), but the onset of MRSA infection outside the hospital, in community-associated settings (CA-MRSA), has been described with increasing frequency. CA-MRSA strains are replacing HA-MRSA strains causing severe infections. A proper evaluation of CA-MRSA prevalence should rely on SCCmec molecular characterization. However, data are limited regarding the genetic background of these pathogens in the CF population. This Italian study investigated the epidemiology, SCCmec type and clinical impact of MRSA strains isolated from nine Italian CF centers. MRSA strains (n=178) were collected from 178 out of 2362 (7.5%) patients. A high prevalence (36%) of CA-MRSA (SCCmec IV) strains was found. The mean patient age at time of the first infection was 11.8 years and 19.2 years in those with CA-MRSA (n=28) and in those with HA-MRSA (n=37) respectively. This difference was statistically significant (p<0.05), suggesting that CA-MRSA strains lead to an earlier onset of the first pulmonary infection compared with HA-MRSA. Decline of pulmonary function was evaluated one year before and one year after MRSA infection. Preliminary data, available on two small groups of patients (6 patients infected exclusively by CA-MRSA and 14 patients infected exclusively by HA-MRSA), showed that FEV 1 decline in CA-MRSA infected patients is more consistent than in HA-MRSA infected patients (-3.67 and +1.07 respectively). Although the observation period is limited and the number of patients studied does not permit statistical analysis, this preliminary observation suggests that more study is needed to evaluate the pathogenic roles of CA-and HA-MRSA. This work was supported by the Italian Cystic Fibrosis Research Foundation (grant FFC#11 2007) with the contribution of Famiglia Riggi.

Objective: The opportunistic mold Aspergillus fumigatus is a frequent colonizer of the respiratory tract in cystic fibrosis (CF) patients. It can cause allergic bronchopulmonary aspergillosis (ABPA), aspergilloma and invasive pulmonary aspergillosis. While antifungal agents in vitro are active against A. fumigatus, in vivo antifungal therapy is often complicated and resistance is observable. The virulence of A. fumigatus is exhibited by production of fungal proteins promoting mycelia growth into the lung parenchyma or by structural features of the conidia that confer resistance to the host antifungal properties. A. fumigatus can grow in multicellular communities on plastic surfaces and on human bronchial epithelia cells (16HBE) and human bronchial epithelia cells with F508del/F508del (CFBE41o-). The proteome and transcriptome of planktonic and biofilmlike grown A. fumigatus mycelium was studied for identification of induced proteins.

Methods: A. fumigatus ATCC #9197 was left to produce biofilm for 24h and 48h. The presence of biofilm on 16HBE and CFBE41o-was demonstrated by using confocal scanning laser microscopy (CSLM). Proteins were isolated and a 2D DIGE gel was performed following MALDI-TOF. In parallel, A. fumigatus cDNA microarrays were hybridized. The results were confirmed by RT-PCR and by detection of proteins in culture supernatants using HPLC analysis.

Results: CSLM images displayed conidia and hyphal structures embedded in matrix formations. A-Alexafluor dyed polysaccharides of the cell wall and of the extracellular matrix (ECM) in the biofilm. Three dimensional constructs of the CSLM pictures displayed biofilm on 16HBE as well as CFBE41o-and proofed viability of the cells after 48h co-incubation. The expression of several proteins changed significantly in biofilm-like grown mycelium. Time-dependent, decreased expression of enzymes of the citric acid cycle and proteins involved in ATP synthesis in the biofilm culture was observed, whereas the glycolytic enzyme enolase and an aldehyde dehydrogenase were upregulated. This points to a developmental stage of the biofilm at 24h demanding energy. At a matured biofilm phase, metabolic activity seems to be reduced. The most striking result was the significant upregulation of proteins involved in the biosynthesis of secondary metabolites. In particular, proteins of the gliotoxin secondary metabolite cluster were induced in biofilm-like grown cultures and a gliotoxin peak was observed after 48h by HPLC analysis.

Conclusions: A. fumigatus is able to produce a biofilm structure in vitro on 16HBE and CFBE41o-. The observed production of secondary metabolites like gliotoxin during biofilm formation in vitro may allow A. fumigatus to survive and persist in vivo during chronic infections.

Cocchi, P. 1 ; Doering, G. 2 ; Braggion, C. 1 ; Taccetti, G. 1 ; Campana, S. 1 

Introduction: Methicillin-resistant Staphylococcus aureus (MRSA) is increasingly recognized in both healthcare-associated (HA-MRSA) and community-associated (CA-MRSA) infections. The molecular characterization of CA-MRSA and HA-MRSA is based on SCCmec typing. SCCmec type I, II and III are responsible for hospital-associated infections, types IV and V are typically defined as CA-MRSA isolates. CA-MRSA are often characterized by a good susceptibility pattern to non beta-lactam antibiotics, and the ability to produce specific toxins. Recently, CA-MRSA are increasingly isolated from hospital settings, causing severe infections and outbreaks. This worrisome trend has been documented also in cystic fibrosis (CF) patients, although the genetic background of MRSA strains involved in their persistent pulmonary infections has been poorly investigated.

Material and methods: Forty-nine CF patients attending the CF center of Florence infected by MRSA have been studied during a 4-year period (1998) (1999) (2000) (2001) (2002) (2003) (2004) (2005) (2006) (2007) (2008) . Fifteen out of 49 patients were persistently infected by MRSA (median length of infection period 5.6 years, range 2-10 years). SCCmec typing was performed as previously described by Oliveira and de Lencastre (Antimicrob Agents Chemother 2002;46:2155).

Results: Seventy-two strains (one strain/year) were collected from these 15 patients and their SCCmec type was determined. Eight out of 15 patients were persistently colonized by the same SCCmec type: 6 were colonized by a HA-MRSA (SCCmec type I) (5 years median period of infection, range 2-9 years), and the other 2 patients were colonized by CA-MRSA SCCmec type IV (median period of infection 2.5 years, range 2-3). The remaining 7 patients are persistently infected by HA-MRSA (SCCmec I) but experience short infection periods (1 or 2 years) due to CA-MRSA (SCCmec IV).

These data highlight the higher persistence of HA-MRSA infections, while CA-MRSA seem not responsible for prolonged infections, but replace HA-MRSA strains for a limited period. The correlation between pulmonary function decline and persistent infections caused by MRSA with different genetic backgrounds should be investigated.

This work was supported by the Italian Cystic Fibrosis Research Foundation (grant FFC#11 2007) with the contribution of Famiglia Riggi.

Nilsson, E. 1 ; Larsson, A. 1 ; Kollberg, H. 2 

Introduction: Pseudomonas aeruginosa (PA) lung infections are main causes of morbidity and mortality in cystic fibrosis (CF). Repeated courses of antibiotics give fewer infections and improved prognosis, but the threat of serious side effects by the use of antibiotics is a reality -especially the rapidly increasing antibiotic resistance. Complements to antibiotics are urgently needed. Chicken antibodies, IgY, have several properties that make them suitable for treatment of infections and there is essentially no risk for development of resistance. We have developed IgY with specific activity against PA, Anti-Pseudomonas IgY.

Aims: To investigate Anti-Pseudomonas IgY as prophylaxis against infections with PA for CF patients and to characterize the antibody treatment.

Results: Anti-Pseudomonas IgY has affinity for PA flagellin, the major component of the flagellum. This is important since the flagellum is required for host invasion and establishment of infection and flagellin induces inflammation. Content of Anti-Pseudomonas IgY: Besides IgY, 26 egg yolk proteins were identified. Some of the proteins have antimicrobial and immunostimulatory effects, which could have a positive additive effect to IgY treatment. Cholesterol levels were low. Currently we are investigating if IgY has an effect on PA biofilm. Preliminary data shows that IgY can bind to the biofilm in a flow cell system.

Seventeen Swedish CF patients have been treated with Anti-Pseudomonas IgY for up to 12 years. The results were compared with a control group of 23 Danish CF patients. Patients treated with IgY had 2.3 PA positive cultures/100 treatment months vs. 7.0 cultures/100 treatment months in the control group (p=0.028), and the time from inclusion to the first recolonization was significantly longer in the IgY-treated group (p=0.012). In the treated group a pair of siblings (12%) became chronically colonized with PA. In the control group seven (30%) have been chronically colonized. No other gram-negative bacteria or fungi have emerged and all PA in the sputa from the IgY-treated patients have been non-mucoid even in the two with several years of chronic infection. Lung function has been preserved and the nutritional status was good at the end of the study. The study has continued for additionally 2.5 years with the same low incidence of PA infections and no new patients have become chronically colonized.

Conclusion: Anti-Pseudomonas IgY is a promising complement in the prevention of PA infections in CF patients, partly explained by the fact that IgY binds to flagellin. The Swedish MPA has given permission to prescribe the drug on licence to specified CF-patients. The European Medicines Agency (EMEA) has granted Anti-Pseudomonas IgY an orphan drug designation with the statement: "…justifications have been provided that avian polyclonal IgY antibody against Pseudomonas aeruginosa may be of significant benefit to those affected by the condition." We are actively searching financial support for a phase III study. Introduction: The Scedosporium apiospermum is the second most frequent filamentous fungus that can be found in patients with cystic fibrosis (CF) after Aspergillus fumigatus. Filamentous fungi (A. fumigatus, S. apiospermum) may contribute to the local inflammatory response and, therefore, to the progressive deterioration of the lung function and under special circumstances those fungi could be pathogenic and invasive.

Aim: To determine the incidence of S. apiospermum colonization in CF patients and explore its possible association with several clinical factors.

Population and method: All 336 CF patients of our department less than 21 years old (156 boys, 180 girls, mean age: 10 years) were surveyed for colonization with S. apiospermum. Sputum cultures collected during routine clinical visits or on admission to the hospital were inoculated on yeast extract-peptone-dextrose agar plates. Colonization was associated with several factors including age, pancreatic function, chronic colonization with Pseudomonas aeruginosa, chronic use of inhaled antibiotics ( >2 years), prolonged use of inhaled steroids (>3 years), and systematic steroid therapy (>3 months).

Results: The overall incidence of S. apiospermum among CF patients was 2%. Yet an incidence of 5.3% was seen in the subgroup of patients with pancreatic insufficiency, chronic colonization with Pseudomonas and chronic use of inhaled antibiotics. Specifically, there was a strong positive corelation with chronic use of inhaled antibiotics (p<0.001) and systematic treatment with steroids (p<0.001). The mean age of the patients at the age of first colonization was 15 years.

Conclusions: The intensification of therapy in CF patients with antibiotics and steroids may facilitate colonization by S. apiospermum. The systematic monitoring for this filamentous fungus at least in a subgroup of patients may be considered.

Spasenovski, T. 1 ; Carrol, M. 2 ; Bruce, K. 1 

The morbidity and mortality arising from respiratory failure due to bacterial infections of the cystic fibrosis (CF) lung is still high. As such, it is important first to better understand these infections and, critically, be able to treat them more effectively. In order to achieve more effective treatment, the manner by which antibiotic susceptibility data are gathered must be considered. Currently, antibiotic assessment is carried out using cells not grown in a biofilm mode of growth -this runs counter to the way in which bacteria typically grow in the CF lung. More than this though, currently, antibiotic susceptibilities are assessed for single species -here, this differs from the complex communities of bacteria that are now known to be present in the CF lung. This study proposes a new model system to assess in vitro efficacy of antibiotic therapy. This model system forms bacterial biofilms in vitro directly from the species present in clinical samples. The degree to which the biofilms that formed resembled the species composition of the original sample could then be assessed. To do this, culture-independent Terminal Restriction Fragment Length Polymorphism (T-RFLP) profiling analysis of 16S rRNA gene PCR products amplified from directly extracted nucleic acids was used. In this way, T-RFLP analysis resolved which bacterial species were present in the original sample as compared to the in vitro formed biofilms. A total of thirty one CF sputum samples, each from a different patient, were studied in this way. The sputa on average contained 4.6±2.3 dominant bacterial species, with the number of species detected decreasing to 4.0±1.6 after five days of bacterial biofilm growth in vitro. Overall similarity of bacterial biofilm communities therefore did not differ significantly over time, with the bacterial content of sputa well reproduced in the biofilm models. Additional measures of biofilm mass and microscopic analysis were also made. Having established model systems in vitro, these were used to study the impact of antibiotics. The central hypothesis here was that through forming models in vitro directly from the multiple bacterial species that these would serve in turn as improved assays of antibiotic susceptibility than conventionally-grown cells. The mean biofilm MICs were 149.8 µg/ml and 53.8 µg/ml for colistin and tobramycin respectively, compared to the MIC values for non-biofilm grown CF sputum samples (76.7 µg/ml and 33.9 µg/ml). This showed that biofilm bacterial communities were significantly more resistant to antibiotic challenge than non-biofilm grown bacteria from the same CF sputum sample. Extensions to this model system also will allow the response of pathogens to antibiotic challenge within this system to be assessed. The potential impact of this strategy in relation to ameliorating existing therapies and for screening for novel treatments is also discussed. cenocepacia strains were tested in competition experiments and directly compared with results obtained from single infections. Paired strains of P. aeruginosa and B. cenocepacia of clinical and environmental origin (RP73/LMG16656, E5/Mex1) were included in the agar beads and challenged in murine lung through intratracheal surgery. After 14 days postinfection, only P. aeruginosa was isolated from lungs of the mice infected with a polymicrobial inoculum (RP73/LMG16656, cfu: 1.8×10 4 /0; Chisquare: p=0.0001; E5/Mex1, 5.4×10 3 /0; p=0.0003). Cytokines (IL-6, MIP2, JE, KC, IL-1β) tested in the murine lung homogenates showed high levels for JE, KC and IL-1β while low levels for IL-6 and MIP2. In particular, IL-1β was significantly higher in mice challenged with polymicrobial inoculum when compared to single infections and independently of bacterial load.

Next, we evaluated the in vitro growth of both free-swimming (planktonic) populations and surface-attached (sessile) cells of B. cenocepacia and P. aeruginosa in single and polymicrobial cultures. Results revealed that the presence of P. aeruginosa negatively influenced the planktonic growth and the biofilm formation of all the B. cenocepacia strains tested, while B. ceno-cepacia only slightly affected the planktonic growth and the biofilm formation of P. aeruginosa strains.

Our in vivo and in vitro results strongly indicate that the two bacterial species may affect each other's virulence and physiology when coexisting. In particular, we found that P. aeruginosa altered the ability of B. cenocepacia to cause chronic infection and to form biofilm while the presence of B. cenocepacia did not affect the virulence and biofilm formation of P. aeruginosa.

( Burkholderia cepacia complex (Bcc) are an important group of pathogens in cystic fibrosis (CF) patients. A subgroup of patients colonized with Bcc can decline rapidly and develop a fatal necrotizing pneumonia. Seventeen species have now been identified within the complex, with B. cenocepacia and B. multivorans being the most clinically relevant. B. multivorans is now the most frequently isolated species in CF patients both in Europe and North America. These bacteria are inherently antibiotic resistant, making Bcc infections difficult to treat with conventional antibiotics. One potential alternative to antibiotic therapy is prevention of adhesion and subsequent lung colonization.

Quantitation of adhesion, and its inhibition, can require cumbersome and time consuming microbiological techniques. We designed a rapid method to quantify B. multivorans attachment to lung epithelial cells (HBE41o-) using Real Time PCR. The RecA gene has been previously used to distinguish between Bcc species using PCR and sequencing. In this method, primers were designed against the RecA gene that were suitable for Real Time PCR (product size 125bp). GAPDH was used as a reference gene against the human cells. The difficulty in this system was distinguishing between bacterial DNA and abundant human DNA which is an issue in many microbiological samples. The method was therefore optimized to exclude non-specific amplification from human DNA. The method was thoroughly validated and shown to correlate well with traditional microbiological techniques and predicted values. Real Time PCR could successfully detect a range from 2 x10 7 cfu to 2 x10 2 cfu, which in the lower range was shown to be more accurate than traditional techniques.

We have previously shown that αβ-linked galactose groups play a role in adhesion as pre-treatment of epithelial cells with α and β-galactosidases decreases adherence and invasion. This method was used to further investigate the role of sugars in the adhesion of B. multivorans to lung epithelial cells by sugar competition assays in vitro. Of the monosaccharides examined, pre-incubation of B. multivorans with galactose, mannose and xylitol all decreased bacterial adherence to lung epithelial cells. Pre-incubation with glucose had no effect. We are also evaluating the potential for a series of novel glycoconjugates to inhibit B. multivorans adhesion with an aim to inhibit adhesion in vivo and ultimately patient lung colonization.

Pulmonary exacerbations (PE) in cystic fibrosis (CF) are a key factor in the irreversible loss of lung function, and as such are strongly linked to morbidity and mortality. During exacerbations, antibiotics are routinely administered in an attempt to reduce bacterial numbers, and in particular, those of Pseudomonas aeruginosa. However, an association between an increase in bacterial numbers in the CF airway, and the occurrence of PE, has not been demonstrated. It has been suggested that PE may be triggered by a change in the mode of growth of key CF pathogens, rather than increase in overall numbers (1) . Determining the relationship between bacterial load and the occurrence of PE is important in the prediction of such events. Further, an understanding of this relationship provides an insight into the manner by which antibiotics have a beneficial impact in the treatment of these clinically important events.

Culture-based diagnostic microbiology is unable to provide accurate quantitative analysis of bacterial load. However, by combining Quantitative PCR (Q-PCR) with PMA treatment, a highly accurate determination of viable bacterial load can be obtained. The aim of this study was to investigate whether a relationship exists between P. aeruginosa, or total bacterial load, and the onset of pulmonary exacerbations in CF.

Sputum samples were collected from 8 adult CF patients and those taken 21, 14, 7 and 0 days prior to PE were retrospectively selected. Samples were treated with propidium monoazide to exclude non-viable bacterial cells. Total DNA was extracted and T-RFLP analysis was performed to determine overall community composition. Total bacterial load was determined by Q-PCR analysis using a universal 16S rRNA protocol, and P. aeruginosa load was determined using a protocol based on the amplification of regA.

Preliminary quantitative PCR data show no significant association between the occurrence of PE and prior increase in either total bacterial, or pseudomonal, load. As such, these data support the model of PE being precipitated by a change in the mode of growth of bacteria within the CF airways, rather than an increase in overall bacterial numbers. The implications for clinical management of PE of these data are discussed. Sinonasal disease is frequent in cystic fibrosis (CF) but symptoms are rarely reported and samples are uncommon. The aim of the study was to analyze clinical and microbiological data in adult CF patients presenting with chronic rhinosinusitis.

Adults with genetically confirmed CF and presenting with chronic rhinosinusitis diagnosed by a senior otorhinolaryngologist on clinical, endoscopic and scanographic criteria were included. Sinus samples were collected during surgery or by middle meatus aspiration and cultured for aerobes on non-selective and selective media (including mannitol-salt agar, and Haemophilus selective medium), and for obligate anaerobes. Results of the last sputum cultures, performed according to published recommendations for CF specimens including selective media for P. aeruginosa (cetrimide agar), S. aureus (mannitol-salt agar) and B. cepacia (OFPBL), and clinical and respiratory function data, were collected.

Nine patients (7 F, 2 M) were included. Their mean age was 26 years (range 22-40 years), and BMI 20 (range 17-24). Pancreatic insufficiency was observed in 8 patients, diabetes mellitus in 6 patients. They all presented with symptomatic chronic rhinosinusitis (nasal obstruction, anosmia, recurrent sinusitis) associated with bronchial infection and decreased lung function. Median FEV1 was 62% (range 25%-102%). Bilateral lung transplantation had been performed in 3 patients 12 to 18 months before symptomatic rhinosinusitis occurred. Sinus samples were collected from paranasal sinus surgery for six patients and under nasal endoscopy for three patients. Two patients presented chronic respiratory colonization with S. aureus, 3 with P. aeruginosa and 4 with both organisms. The compared bacteriology of sinus and sputum is shown in Table 1 . None of the sinus cultures yielded strict anaerobes.

The compared antibiotic susceptibility profiles of sinusal and respiratory isolates of S. aureus were identical in all patients. Several morphotypes exhibiting different antibiotic susceptibility profiles were identified in the sputum cultures of 4 of the 6 sinus positive patients, whereas a single morphotype was identified in sinus cultures. Only one of the P. aeruginosa strains recovered from sinus samples was mucoid, whereas mucoid strains were recovered from the sputum of all patients. Finally, in the patient who was intermittently colonized with S. maltophilia, the antibiotic susceptibility profile of the sinusal isolate was markedly different from that of the respiratory isolates. After surgical treatment, an individual functional benefit was observed in the 6 patients and the number of antibiotic courses decreased.

This preliminary study on a small number of patients suggests that sinusal flora may act as a reservoir for pulmonary infection in CF adults. Sputum surveillance for Non-Tuberculous Mycobacteria (NTM) was undertaken at least once yearly (where possible) and additionally when suspicious features were detected on CT scans (such as inflammatory nodules, cavities or new "tree in bud" change) or if patients failed to respond to conventional antibiotics during infective exacerbations.

Between 2004-2008 the annual incidence in new NTM positive sputum cultures increased successively from 1.4% (3/217) in 2004-5, 2.3% (5/221) in 2005-6, 3.1% (7/228) in 2006-7 and 4.3% (10/230) in 2007-8 . Over this period, the most frequently isolated species were M. abscessus/chelonae (10/25), M. avium/intracellulare (5/25) and M. kansasii (3/25) . Ten patients during this time period were diagnosed with NTM disease of which 8 had M. abscessus.

We have developed a treatment algorithm for M. abscessus (believed to be the NTM most difficult to treat) involving 2-3 weeks induction therapy with 3/4 intravenous antibiotics followed by maintenance therapy with 3/4 oral and 1/2 nebulised antibiotics with or without subcutaneous interferon gamma. Exacerbations are treated with additional intravenous antibiotics for 2-3 weeks followed by augmentation of maintenance therapy.

The total number of patients treated with M. abscessus/chelonae (including those referred to our unit with untreated disease and those who commenced treatment prior to 2004) is 13 of which: 4 remain clear of NTM for at least 12 months after discontinuing treatment; 5 are still on treatment but have become culture negative for at least 6 months while 4 patients continue to culture M. abscessus despite maximum medical therapy.

We conclude that the incidence of NTM disease is increasing in adults with cystic fibrosis, that M. abscessus is the major pathgenic species and that successful treatment can be achieved in at least some patients. Background: Early anti-Pseudomonas aeruginosa eradication treatment in cystic fibrosis patients is an efficacious method to delay chronic infection due to this pathogen. Regardless of the type of antibiotics selected to treat P. aeruginosa, treatment efficacy is high (85%). Definition of successful eradication therapy was based on clinical (UK, CF Trust Guidelines) and molecular criteria carrying out genotyping of the first isolated P. aeruginosa strain and the strains isolated from successive colonizations. In fact the molecular approach can be used to distinguish between re-growth of the same strain (unsuccessful treatment), suppressed but not eradicated, and recolonization by new strains (successful treatment).

Aims: In order to evaluate the influence of repeated eradication treatment on antibiotic susceptibility of P. aeruginosa MIC of all strains were compared between initial and subsequent re-isolations in patients where eradication was successful and in those where treatment had failed.

Methods: Molecular study of bacterial isolates was performed using RAPD-PCR. MIC 50 and MIC90, MIC's at which 50% and 90% of isolates are inhibited, of 46 isolates of P. aeruginosa from 18 cystic fibrosis patients in follow-up in our Center over a 10-year period were evaluated by E-test for the following antimicrobials: ciprofloxacin, tobramycin, ticarcillin-clavulanate, ceftazidime and meropenem. Eleven strains isolated from the first colonization episodes, 28 strains isolated from re-colonization episodes after successful eradication treatment, and 7 strains isolated from unsuccessful eradication episodes were studied.

Results: Susceptible P. aeruginosa strains are shown in Table 1 . There were no statistically significant differences in antimicrobial sensitivity of P. aeruginosa isolates in the 3 groups (one way ANOVA).

Conclusion: Repeated eradication therapy of successive episodes of newly acquired organisms did not affect overall antibiotic resistance rates over the period of the study of P. aeruginosa strains isolated both after a successful eradication treatment (newly acquired genotype) and after an unsuccessful therapy (re-growth of the same genotype). These results support the use of the eradication treatment in all cases of new acquisition of successive P. aeruginosa strains.

Prevalence of P. aeruginosa susceptible strains isolated during the first colonizations, the re-colonization episodes and the re-growth episodes. Microbial pathogens cause damage by the production of proteases and toxins harmful to lung tissues and by interaction with elements of the innate immune response. Early Pseudomonas aeruginosa infection is typified by the production of quorum sensing (QS)-regulated factors involved in virulence and biofilm maturation which are essential for or augment pathogenicity. Cells cultured from patient's sputum during the early, acute phase exhibit a wild-type QS-active phenotype. If infection is not eliminated, selective pressures in the lung lead to the cumulative acquisition of selectively advantageous mutations. The loss of aggressive virulence factors during progression from acute to chronic phase is well documented (Smith et al. Proc Natl Acad Sci USA. 2006 30; 103(22) :8487-92) and is frequently attributable to disruption of the QS communication systems (D'Argenio et al. Mol Microbiol. 2007 64(2) :512-33). Despite loss of QS-regulated factors infection persists and lung damage continues, presumably due to alternative modes of pathogenicity. This study aims to determine virulence and pathogenicity factors associated with chronic rather than acute infection.

We have used both transcriptomic and proteomic technologies to perform in-depth comparative analysis of two clinical P. aeruginosa isolates derived from CF lung infections. One isolate was QS-active, i.e. typical of early, acute infection, and the other a QS-compromised isolate, i.e. typical of adapted, chronic infection. Comparison of in vitro biofilm growth of each isolate confirmed the absence/reduction in the chronic isolate of proteins involved in the las, rhl and PQS signalling systems and many QS-associated virulence factors. A number of factors known to contribute to virulence and pathogenicity in animal infection models were detected as specific to or up-regulated in the chronic, QS-compromised isolate, including isocitrate lyase (11 fold up-regulated) which has previously been shown to be a marker of chronic phase virulence. Proteins with virulence functions in other pathogenic bacteria not yet recognised in P. aeruginosa, were also observed. These represent candidate chronic pathogenicity factors and exhibit functions implicated in LPS modification (5 proteins, 6-18 fold up-regulated), oxidative stress tolerance (7 proteins, 5-18 fold up-regulated) , intracellular survival (6 proteins, 5-7 fold up-regulated) and 2 putative hemolysins (11-22 fold up-regulated).

Together the data suggest that adaptation of P. aeruginosa during chronic infection of the CF lung focuses on evasion of the host innate immune response and may include an increasing ability to survive intracellularly as has been reported for Burkholderia. The prospect of intracellular survival of P. aeruginosa has important implications for treatment of persistent infections. Biofilms provide a reservoir of potentially infectious microorganisms, that are resistant to antimicrobial agents. Their importance in chronic inflammatory conditions, such as cystic fibrosis (CF) is increasingly recognised. Fluorescent in situ hybridisation (FISH) is a technique that can be used to visualize the 3-dimensional structure of biofilms and to identify specific bacterial populations within the biofilm itself, including those encountered in the cystic fibrosis lung. In this study, FISH utilising Peptide Nucleic Acid (PNA) probes, was used in combination with confocal laser scanning microscopy (CLSM) to detect and characterise the spatial distribution of biofilm-forming bacteria which predominate within human chronic infections. In vitro biofilms were prepared using the Constant Depth Film Fer-menter (CDFF) and a reconstituted human epidermis (RHE) model. The specificity of PNA probes for Pseudomonas aeruginosa and universal bacterial detection was initially determined using planktonic bacterial preparations. Subsequently, biofilms comprising target and non-target species were prepared on PTFE discs in a constant depth film fermenter, and on reconstituted human epithelium (SkinEthic Laboratories). Biofilms were washed with PBS and PNA probes (100-400nM) applied for 1 h at 55∞C. Sections of these stained biofilms were then viewed by CLSM. The specificity of the PNA probes was confirmed using mixed preparations of planktonic bacteria and multiplex PNA probing. The identification of target bacteria within the biofilms was also evident, confirming the suitability of the approach for biofilm analysis. CLSM revealed clustering of individual species within the biofilm, with limited co-localisation of mixed species; defined aggregates of bacteria being evident in the dermis of CVLUs. The development of this standardized procedure makes available an assay for simultaneous identification of single or multi-species bacterial populations using distinctly labelled PNA probes. The protocol presented gives a reliable basis to design and perform PNA FISH analysis on both bacterial biofilms and infected tissue and offers an approach to assess targeted biofilm disruption strategies in vivo.

Longterm use of the macrolide, azithromycin, has shown a significant clinical improvement for patients with cystic fibrosis (CF). However, its longterm effect on the susceptibility of the commensal flora within the CF airways to macrolides has not yet been examined. The aim of this study was therefore to determine the level of erythromycin resistance in naturally colonising viridans group streptococci (VGS) and pneumococci, as well as to characterize macrolide resistance gene determinants via sequence analysis. Streptococci (190) were isolated from 54 sputum specimens from 41 adult CF patients, 35 who were on longterm oral azithromycin therapy for at least the previous seven months. Macrolide resistance was examined by standard disk diffusion assay and molecularly for the presence of erm(B) determinant, as well as mef(A/E). Among the isolates tested, the majority of them (128; 67.7%) were resistant to erythromycin, whilst a further 37 (19.4%) had reduced susceptibility, leaving only 25 (13.2%) isolates sensitive. These results are well correlated to the presence of macrolide resistance determinants, as demonstrated by possession of the erm(B) gene 29/190 (15.3%), the mef(A/E) gene 129/190 (67.9%) and both erm(B) + mef(A/E) genes simultaneously 14/190 (7.4%). These results indicate that genotypic resistance for macrolides is common in VGS in adult CF patients, with efflux being over three times more frequent as a resistance mechanism than rRNA methylation, in comparison. Chronic treatment with azithromycin in patients with CF may reduce antibiotic susceptibility in the commensal streptococcal flora, where these organisms may potentially act as a reservoir of macrolide antibiotic resistance gene determinants for newly acquired and antibiotic sensitive pathogens. Reid, A.; McCaughan, J.; Jenkins, L.; Leavy, A. Paediatric, Royal Belfast Hospital for Sick Children, Belfast, United Kingdom Background: Pseudomonas aeruginosa (PA) remains an important common pathogen in the cystic fibrosis (CF) lung. Chronic PA infection is associated with an increased morbidity and mortality in the CF population. Over the last 15 years there has been major emphasis on early identification and eradication of PA in an attempt to stop or slow progression to chronic infection. Since the early 1990s we have adopted various modifications of the Danish Protocol for PA eradication. In 2004 we also introduced segregation of all inpatients and outpatients, not just those colonised with B. cepacia or MRSA.

Aim: The aim of this study was to establish if our current interventions to prevent or eradicate PA have led to a reduction in the incidence of chronic PA infection within our screened CF population.

Methods: Data was retrospectively analysed from our dedicated paediatric bacteriological database from 1995-2008. Each patient's yearly bacteriological status was established using the Leeds criteria to identify chronic or intermittent PA infection.

Results: Patient numbers each year remained around 200 (183-235), with an age range from 0-18 years in a screened population. Overall PA was isolated from between 30 -40% of our patients in each calendar year. During this period the incidence of intermittent PA has increased from 6% in 1996 to 22% in 2008, but the incidence of chronic infection has dropped from 30% in the late 1990s to 15% in 2008. Phenotypically the PA isolates found in most intermittent infections were pigmented, non-mucoid and antibiotic sensitive, suggesting acquisition from the environment.

Conclusions: Although the overall rate of PA has remained relatively unchanged the rate of chronic infection has decreased dramatically. Indications are that environmental acquisition of PA continues to be significant but current interventions appear to be successful in reducing progression to chronic PA. The purpose of our study has been to estimate the effect of the prophylaxis with Pal on the time of first PA infection. We also aimed to determine the effect of Pal treatment on hospitalization for acute respiratory illness in young children with CF during the first RSV season that followed diagnosis of CF.

Eighty-four children with CF,diagnosed by neonatal screening, born between 26 January 2000 and 26 May 2007, were observed (33 treated and 51 non-treated with Pal-control group). All parents provided written informed consent before treatment. We confronted the age of patients in the moment of the first infection with PA, the mean number of the routine visits and the mean number of hospitalizations for APE in the first year after the treatment. The results were statistically elaborated using t-test (for comparing the difference of number of visits and for comparing number of hospitalizations in the first year of life between two groups) and linear regression considering difference in sex (for mean age of the first PA infection difference).

The treated patients had more routine visits for year, owing to Pal administration: 14.28 vs 9.73 (p=0.0017) but not significantly less hospital-ization for APE in the first year of life: 0.36 vs 0.65 (p=0.271). Thirty-three of our patients had the first infection with PA (15 treated and 18 non treated with Pal). The Pal treated population had the first PA infection later then non-treated population for 99.51 days, but without significant difference (p=0.292).

In conclusion: in our study immunization with Pal does not affect significantly time of the first PA infection. These data will be revaluated with our further prospective study. 1986;153:902) . During the normal physiological, disease-free state and in the absence of inflammation, the anti-proteinase activity of A1AT outweighs the proteinase burden, hence preventing hNE-mediated damage to airway tissue and local defences. Previously Morihara et al. (J Biochem 1984; 95:795) demonstrated that Pseudomonas elastase (LasB) degrades A1AT within the exposed reactive site loop or RSL domain of the A1AT molecule, the region that complexes with hNE. In the current study we investigated if secreted bacteria-derived proteinases from Burkholderia multivorans, Burkholderia cenocepacia and Pseudomonas aeruginosa strains were able to degrade A1AT, and furthermore, if such degradation affects the inhibitory activity of this proteinase inhibitor against hNE.

Methods: Plasma-derived A1AT (A1AT) was incubated with relevant cell-free bacterial supernatants from the three bacterial species for 24 hours at 37∞C. The hydrolysis of the PD-A1AT molecule was assessed by SDS-PAGE followed by Coomassie blue staining. To assess if the degraded A1AT retained its functionality, the ability of the molecule to complex with recombinant human elastase (rhNE) was investigated by SDS-PAGE and was assessed by the ability of rhNE to hydrolyse the substrate N-Methoxysuccinyl-Ala-Ala-Pro-Val-7-amido-4-methyl-coumarin (AMC) in a fluorogenic enzyme activity assay.

Results: Incubation of PD-A1AT with bacteria-derived proteinases belonging to B. multivorans demonstrated that three isolates (C5393, C1962, CF-A1-1) produced proteinases that were able to hydrolyse the A1AT molecule. Degradation of A1AT by proteinases derived from C5393 and CF-A1-1 demonstrated a ~40% loss in inhibitory ability against rhNE. The single remaining B. multivorans (C1962) and both B. cenocepacia strains (J2315, J415) did not impair the ability of A1AT to inhibit rhNE activity, even though a limited degradation of the A1AT molecule was observed in all cases. Degradation of the A1AT molecule was also observed by proteinases derived from three P. aeruginosa strains (PA0048, PA0219, PA0239). Assessment of the truncated form of the A1AT showed that it lost its ability to complex with rhNE, which was further evidenced by a near 50% reduction in its inhibitory activity towards rhNE.

Conclusion: A number of bacterial isolates belonging to B. multivorans and P. aeruginosa are able to degrade A1AT resulting in a loss of a small peptide of ~3 -5 kDa. This results in a significant loss of A1AT inhibitory activity against hNE. These results demonstrate that bacterial proteinases are potentially significant contributors to the observed proteinase / anti-proteinase imbalance within the CF lung. Here, we analysed, at chemical and biological level the two major components of the P. aeruginosa cell wall, the LPS and peptidoglycan (PGN), isolated from three clonal strains of one P. aeruginosa lineage. P. aeruginosa strains were isolated at the onset of infection and after 7.5 years of chronic colonization from a CF patient with a severe course of the airways infection. Chemical structure by MS spectrometry defined a lipid A that is variably penta, hexa or hepta-acylated and associated with the different stages of CF infection. According with these chemical changes, the sequence analysis of pagL, coding for Lipid A 3-O-deacylase, interestingly revealed an adaptive mutation in a non-mucoid late strain.

As for the PGN, diversity consisted in different distribution of canonical monomeric and dimeric species. When tested in human cells including those of CF origin, the strong inflammatory response induced by P. aeruginosa LPS and PGN isolated at early stage of infection was attenuated at late stage. Significantly higher NF-κB activation, IL-8 expression and production were detected in HEK 293-hTLR4/MD2-CD14 after stimulation with LPS and HEK293hNod1 after stimulation with PGN of early strain when compared to late strains. Similar results were obtained in IB3-1 cells of CF origin.

Next, to test whether the differences in the lipidA structures could affect the inflammatory response in vivo, we analyzed leukocyte recruitment in the BALF of C57Bl/6 mice exposed to different LPS structures of clinical strains by means of nasal instillation. Neutrophil profile showed striking differences in total differential cell counts, while there was no significant difference in monocytes and lymphocytes number. Significant higher recruitment of neutrophils was observed in mice exposed to LPS from early strain in comparison to those treated with late strains. Cytokines levels, tested in murine lung homogenates, showed higher MIP-2 levels for mice treated with early LPS than late LPS. Similar trends were obtained with KC and IL-1 beta. Histopathological analysis of lung tissue sections confirmed differences between early and late LPS.

Our results suggest that the chemical changes in both of P. aeruginosa LPS and PGN follow the host innate immune system evasion and endorse bacterial pathogenesis.

Supported by the Italian CF Research Foundation (FFC#8/2006 and FFC#8/2007). 3, 4 ; Corbett, C.R. 1, 2 1. National Microbiology Laboratory, Public Health Agency of Canada, Winnipeg, MB, Canada; 2. Medical Microbiology, University Of Manitoba, Winnipeg, MB, Canada; 3. Department of Microbiology and Infectious Diseases, University Of Calgary, Calgary, AB, Canada; 4. Department of Biochemistry and Molecular Biology, University Of Calgary, Calgary, AB, Canada Cystic fibrosis (CF) is the most common fatal genetic disease affecting young Caucasians. CF is a multi-organ disease that primarily affects the lungs and digestive system, with the majority of mortalities resulting from pulmonary failure due to repeated pulmonary exacerbations. Recently the Streptococcus milleri group (SMG: S. anginosus, S. constellatus and S. intermedius) has been implicated in pulmonary exacerbations of CF patients due to the development of a SMG selective media. This media has allowed for the detection of SMG and shows that it is an under-reported respiratory pathogen. Our aim was to develop a real-time PCR assay to complement the detection of SMG within diagnostic samples. The cpn60 gene was chosen as a target based on multiple sequence alignments showing conserved regions within SMG. These conserved regions were used to develop primers and probes that were selective for SMG. The Roche Lightcycler 2.0 was used for real-time PCR assays, along with Roche Hybridization probe real-time PCR technologies. Streptococcus constellatus, S. intermedius, and S. anginosus ATCC type-strains can be amplified using a single hybridization probe set. Clinical strain analysis showed that speciation based on melting curve analysis allowed for the majority S. constellatus and S. intermedius isolates to be correctly identified and most of the S. anginosus strains. To improve our ability to detect SMG in complex polymicrobial specimens, two 16s rRNA real-time PCR assays were developed targeting the 16s rRNA gene. The 16s_SA assay is specific for S. anginosus, while the 16s_SCI assay is specific for S. constellatus and S. intermedius. These assays can detect less than 10 genome equivalents for the target strains in pure culture and >10 5 genome equivalents in complex polymicrobial specimens. This sensitive quantitative molecular test will allow rapid identification of SMG in sputum samples. 

The Streptococcus milleri Group (SMG) have been described as commensal members of the human oral microbiota that have the ability to cause severe invasive infections including brain and liver abscesses. Recently members of the SMG group have been shown to be important pathogens in pulmonary exacerbations of cystic fibrosis (CF) patients. Currently, there is little information on the genetic composition of SMG and factors contributing to virulence. We conducted whole genome sequence analysis of two isolates of each species Streptococcus anginosus, Streptococcus constellatus and Streptococcus intermedius, that comprise the SMG group. For each species a representative strain from CF airways (numerically dominant isolate during acute exacerbation) and invasive isolates (from other body sites) were analyzed to study the genetic relationships of the pathogens. Pyrosequencing, using the GS FLX platform, was used for whole genome sequence generation. Sequence was arranged using the Staden suite of pro-grams and gaps were filled using a combination of fosmid library development and primer walking. The genomes were then run through the NML Bioinformatics core genome analysis pipeline, including analysis with GenDB suite of programs for automated annotation, checked for potential frameshifts, and finally manually annotated to ensure proper start calls for genes and appropriate detection of regions for genes with lower known homologies. Any sequence deviations including SNPs, length of homopolymers, and insertions or deletions were checked using PCR and traditional Sanger sequencing.

Streptococcus constellatus was the smallest genome with 1663 automated gene calls and a genome size of 1935392 bp and a 38.18% G+C content, S. intermedius had 1666 gene calls a genome size of 1996181 bp and 37.56 G+C content, while S. anginosus was larger with 2024 gene calls and a genome size of 2233408 Mbp and 38.25 G+C content. The sizes of these genomes are within the range of already sequenced Streptococcus genomes and the G+C content is within the previously predicted range for SMG (37-41%). The availability of sequence data for these organisms will allow indepth study of this pathogen such as development of specific PCR diagnostic assays, insight into potential virulence factors and novel information regarding the biology of the organisms. The airways of patients with cystic fibrosis (CF) are typically infected with multiple pathogens, including but not limited to Pseudomonas aeruginosa (PA), Staphylococcus aureus, Stenotrophomonas maltophilia, Achromobacter xylosoxidans, Burkholderia spp. and Aspergillus ssp. There is concern that an alteration in the bacterial microenvironment by an antibiotic can promote growth of untreated or resistant pathogens. AZLI is an anti-pseudomonal antibiotic with a lysine excipient that has significantly reduced mean sputum PA density in two 28-day course multicenter, randomized, Mortality in cystic fibrosis (CF) is most commonly a result of massive bacterial infection in the lungs due to reduced mucociliary clearance. Although the most common bacterial species found in CF lungs are classic opportunists like Pseudomonas and Burkholderia, most infections consist of a multispecies community. Bacterial communities can display emergent properties due to the complex interactions within the community and with the host tissue, and the impact of these interactions on the overall pathology of CF is not fully understood. To examine these complex bacterial infections and interactions in a CF-like environment, we have developed a novel technique for the installation of a mucus simulant that closely mimics the viscoelastic properties of CF mucus. Low melting point agarose (LMPA) maintains viscoelastic properties similar to water at high temperature (42 o C) but forms a viscoelastic gel upon cooling to 37 o C. C57/BL6J mice were infected intratracheally with varying doses of Pseudomonas aeruginosa (1 x 10 3 -1 x 10 7 ) with low LMPA or saline and examined 72h later. Challenges with agarose resulted in decreased survival of the mice, with nearly 70% dying within 48 hours when dosed with more than 1 x 10 5 bacteria. Similar numbers of bacteria suspended in saline were well tolerated and all survived. Bacteria (10 5 ) suspended in saline were more than 90% cleared by 72h, but when suspended in agarose were maintained at the inoculum density or higher. Across the dose response curve, LMPA-Ps challenged animals exhibited increased total cell and neutrophillic infiltration of the airways, along with increased expression of key inflammatory cytokines compared to saline-Ps challenged mice. Histologic examination confirmed restriction of infection and inflammation to airways of the LMPA-Ps dosed animals compared to control, and hypoxic regions in the airway lumen. Low volume LMPA-Ps challenge appears to be a good technique to develop chronic bacterial airways infections with hypoxic niches. CF Center, Fondazione IRCCS Ospedale Policlinico, Mangiagalli, Regina Elena, Milan, Italy; 2. CF Centre and Microbiology Lab, Fond. IRCCS, Osp. M. Policlinico, MaRE, milano, Italy Mycobacterium abscessus (MA) and Mycobacterium chelonae (MC), two species of rapidly growing mycobacteria (RGM), have been isolated increasingly often from the respiratory tracts in cystic fibrosis (CF) population and implicated in responsibility for CF chronic lung disease.

It is not known whether these organisms are trasmitted from person to person or acquired from environmental sources.

Aim: 1) Identification to the species levels of MA and MC using a commercial assay based on the reverse hybridisation of PCR products. 2) To determine an hypothetical source of infection studying genetic fingerprinting of MA and MC, obtained by rep-PCR (Repetitive Extragenic Palindromic) method.

Methods and Patients: Starting from January 1, 2008, 2500 sputum samples from 625 CF patients followed by Milan CF centre, were routinely processed and investigated, according to microbiological standard procedure, every 3 months over a one year period.

All supposed strains of MA and MC isolated were stored at -80∞C. Genetic identification of strains was obtained by means of DNA strip assay (GenoType Mycobacterium CM; Hain, Lifescience, Nehren, Germany). After this first step, all identified strains were subjected to rep-PCR (Diver-siLab Mycobacterium Kit) to compare their DNA fingerprinting using DiversiLab System (BioMerieux S.p.A.).

Results: Twenty two CF patients (mean age±SD=23.5±9,97 years) were found positive for RGM (prevalence 5,3%). Twenty patients were colonised by MA and 2 by MC. The 22 strains, one for each positive patient, were tested by rep-PCR. Eight of them showed no genetic correlation with any of the other isolates tested. Fourteen fell within 5 clusters:

-Cluster 1 and 4 each constituted by 2 similar strains (respectively 92% and 93% similarity); -Cluster 3 and 5 each constituted by 2 indistinguishable strains (respectively 99% and 96% similarity); -Cluster 2 constituted by 5 indistinguishable strains and 2 similar strains (respectively 99% and 92% similarity).

Conclusions: 1) As MA seem to have a more pathogenic effect than MC on CF lung, an accurate identification is important to establish an appropriate therapy.

2) The automated rep-PCR based DNA typing platform, DiversiLab System, represents a useful tool for molecular epidemiology in clinical laboratory. It is also useful in surveillance of circulating pathogens and in disease outbreak.

3) Up to now, our data suggest a possible person-to-person transmission of MA.

Special thanks to Dr. Alessandro Foschini of BioMerieux Italia S.p.A., our ever present expert consultant. Repetitive Extragenic Palindromic (Rep)-PCR is a molecular biology method, standardised and easy to use, to assert genetic correlation among different bacterial strains.

Aim: To establish if the reappearance of Pseudomonas aeruginosa (PA) in CF patients, who underwent eradication therapy, is due to a newly acquired strain or is the same as the first acquisition.

Method: Starting from January 2008, 38 pts followed at the Milan CF centre were enrolled at the time of their first lung infection by PA (T0). These patients were treated for 28 days with two different antibiotics (inhaled colistin/oral ciprofloxacin and inhaled tobramycin/oral ciprofloxacin) and subjected to 3 microbiological controls during the following 6 months. All strains isolated from sputum culture at T0 and the ones eventually acquired during 6 months after antibiotic therapy (T1) were stored at -80∞C. Isolated strains were successively subjected to rep-PCR (DiversiLab Pseudomonas kit) to compare their DNA fingerprinting using DiversiLab System (BioMerieux S.p.A.).

Results: Eight of the 38 pts were found positive for PA during 6 months after therapy. We compared the genetic fingerprinting, obtained with rep-PCR, of the strains isolated at T0 and T1 of each pts.

In 5 pts, strains recovered at T1 were indistinguishable from the strains isolated at T0 and reinfection occurred within the first 3 months after antibiotic therapy. In the remaining 3 patients, strains of the first acquisition were found to be different from the ones isolated at T1 and the patients were found positive for PA between the third and the sixth month after therapy.

Conclusions: The ability of rep-PCR to distinguish isolates at the clonal level indicates that it can be a useful tool in monitoring the efficacy of antibiotic therapy in CF patients with early PA infection. This method could be also valuable for PA infection control intervention.

Special thanks to Dr. Alessandro Foschini of BioMerieux Italia S.p.A., our ever present expert consultant. 

IgY antibodies derived from the yolk of immunised hens have been used for multiple applications most commonly in the gastrointestinal tract, for example gastroenteritis and oral bacterial disease. Specific Pseudomonas aeruginosa (Pa) IgY has been shown to delay/ prevent infection with Pa in children with cystic fibrosis (CF) when taken daily as a gargle. The mechanism behind such an effect is not understood. We hypothesized that the IgY antibodies may be inhibiting the adherence of organisms to cells of the oropharynx and tested this hypothesis with a modified bacterial adherence assay. Using similar methodology, we have previously reported increased adherence of Pa to the respiratory epithelium of CF patients, which could be reduced with CFTR gene transfer or antibodies to the putative receptor, asialoGM1.

In this study, we obtained cells from the buccal mucosa of healthy volunteers by scraping with a sterile spatula. Three serotypes of bacteria (03, 06, 011) to which hens had been immunised were used. After suspension in PBS, concentration was standardised using spectrophotometry. Two experimental conditions were tested: a) exposure of bacteria to IgY (final concentration 175 mcg/ml; 1 hour at 37 o C) prior to incubation of Pa with buccal cells and b) exposure of cells to IgY prior to incubation with the bacteria. The latter was thought more likely to mimic conditions in vivo. For each sample, purified IgY from non-immunised hens was used as a control and a third group was exposed to PBS only. Following incubation of the cells with bacteria, non-adherent organisms were removed by spinning the buccal cells through a percoll gradient. Cells were then cytospun onto thermanox coverslips and processed for scanning electron microscopy. Adherent bacteria were quantified on coded samples.

For two of the bacterial serotypes (06, 011), specific IgY significantly increased adherence to buccal cells (between 5 and 20 fold) compared with either non-specific IgY or PBS (which were not different from each other). Interestingly, a similar effect was observed when either bacteria or buccal cells were first exposed to the IgY. For serotype 03, there were no differences between groups for either condition.

These findings clearly show that specific Pa IgY has an effect on Pa. However, the increased adherence mediated by IgY was surprising and it contradicts earlier findings of a decreased adherence to epithelial cells of healthy individuals obtained from skin or the external auditory canal. This discrepancy might be explained either by differences in cell source or differences in techniques. A plausible explanation of the clinically demonstrated efficacy would be that bacteria are anchored in the oropharynx after gargling with specific Pa IgY and cleared by local defence mechanisms rather than being inhaled into the lower airway where infection can be established. 1 1. Adult Cystic Fibrosis Centre, Cork University Hospital, Cork, Ireland; 2. Otorhinolaryngological Department, South Infirmary Victoria University Hospital, Cork, Ireland Introduction: Chronic rhinosinusitis (CRS) is a well-recognized clinical syndrome affecting patients with cystic fibrosis (CF) and has been implicated in recurrent infective pulmonary exacerbations. Self-reported symptoms in the CF population are low, despite a high prevalence of radiological and clinical sino-nasal disease. Few studies have addressed the implications of unrecognized upper airways disease on clinical phenotype in adult CF.

Objective: A prospective study to determine the incidence of symptomatic CRS in adult CF patients, and to stage the severity of CRS using validated endoscopic and CT scoring systems. A secondary objective was to establish the effects of CRS on clinical phenotype in adult CF.

Consecutive adult patients with no prior history of nasal sinus surgery, attending the Cork Adult Cystic Fibrosis Centre undertook a symptom-based validated CRS questionnaire, flexible nasal endoscopy and CT of nasal and paranasal sinuses. The endoscopic Lund-Kennedy Scoring system (0-12) graded on the presence of nasal polyposis, nasal mucosal edema and nasal secretions was calculated for each patient. A modified Lund-MacKay score for CTs of nasal and paranasal sinuses was also determined. Spearman and Mann-Whitney U tests were performed to correlate Lund-Kennedy Score with markers of disease severity.

Results: To date 47 patients have met inclusion criteria. Flexible nasal endoscopy demonstrated CRS in 59.6% of patients versus 37.5% of patients by symptomatology score (p=0.08). Symptomatic patients had a significantly higher Lund-Kennedy score (4.83 vs 2.62, p=0.003) compared to asymptomatic patients. There was no significant association between Lund-Kennedy score and class mutation, age, frequency of infective pulmonary exacerbation rate over the previous 4 years or lung function. In a sub-group CT has been performed in 15 patients to date: 93% of patients were found to have hypoplastic frontal sinuses; 87.5% of the asymptomatic patients had Lund-MacKay CT scores demonstrating CRS. There is a significant strong correlation between Lund-Kennedy nasal endoscopy scoring and Lund-MacKay CT scoring (r=0.564, p=0.044) for the detection of CRS.

Conclusion: CRS is under-diagnosed in adult CF patients. We propose that this is due to unrecognized symptoms by patients. Endoscopy and CT offer more sensitive methods than questionnaire to detect CRS in adult CF patients, and may correlate with clinical phenotype. Further radiological imaging is necessary to assess this.

Voase, N.W. 1, 4 ; Davies, G. 1, 4 ; Reid, P.A. 2, 4 ; Macleod, K.A. 3, 4 ; Dewar, M.H. 2, 4 ; Bell, N.J. 2, 4 ; Saunders, C. 1, 4 ; Greening, A.P. 2, 4 ; Cunningham, S. 3, 4 ; Innes, J.A. 2, 4 ; Alton, E.W. 1, 4 ; Davies, J.C. 1, 4 1. Gene Therapy, Imperial College London & Royal Brompton Hospital, London, United Kingdom; 2. Western General Hospital, Edinburgh, United Kingdom; 3. Sick Childrens Hospital, Edinburgh, United Kingdom; 4. UK CF Gene Therapy Consortium, Edinburgh, London, Oxford, United Kingdom Microbiological surveillance is the mainstay of cystic fibrosis (CF) clinical monitoring. Prompt recognition allows tailored treatment and may prevent chronic infection which adversely impacts outcomes. Many patients, most commonly children and adults with milder stages of lung disease cannot expectorate sputum spontaneously. Alternatives include throat swabs, cough swab (CS) or cough plates but there are concerns about the sensitivity of these.

The UK CF Gene Therapy Run-in study involves patients from London and Edinburgh aged 10 years and over. Patients are seen at regular intervals (> 3 months) during periods of clinical stability for measurement of a panel of biomarkers which are being assessed as end-points for our gene therapy clinical programme. Any patient that is either non-productive or produces <1.1ml sputum (required for all assays) undergoes sputum induction with 3.5-7% NaCl. These subjects also undergo CS culture at the same visit. This affords us a unique opportunity to address the success and utility of sputum induction in those patients where insufficient spontaneous sputum is produced.

A total of 534 Run-In study visits have been completed across the two sites, involving 191 patients. Expectoration status divides patients into 3 groups: a) sufficient spontaneous sputum produced (52% of visits; group 1); b) some produced but induction also required (9% of visits; group 2); c) no spontaneous sputum, induction required (39% of visits; of these sputum was obtained on 66% of occasions, group 3).

In the group unable to expectorate but producing sputum after induction (group 3) on whom a full dataset is available (n=41), cultures from induced sputum (IS) and CS were fully concordant in 41% (n=17: 10 both negative; 7 both positive for the same organisms). In a further 15% (n=6), both samples were positive but incompletely concordant: 4 had an additional organism (S. maltophilia, A. fumigatus, S.aureus, H. influenzae) in IS which was not detected on CS; 2 patients had only Af in sputum, but had Pa (n=1) and Sa, Hi and Sm (n=1) on cough swab. Importantly samples in 44% of cases (n=18) were completely discordant and in all of these a significant pathogen was grown only from IS whilst CS culture was negative.

Similarly, in group 2 (mixed spontaneous and IS; n=16) 56% (n=9) had a negative cough swab despite one or more significant pathogens in sputum; only one patient had a negative sputum culture with a positive CS (P. stutzeri).

Culture of induced sputum provides additional, clinically-relevant information compared with cough swab in non/poorly-expectorating CF patients. We suggest that IS should be considered more routinely in the clinical setting; its use could save patients from more invasive investigation such as bronchoalveolar lavage and could result in the earlier and more rational treatment of lower airway infection. Agmatine is the decarboxylation product of arginine and a number of bacteria have devoted enzymatic pathways for its metabolism. Pseudomonas aeruginosa harbors the aguBA operon that is upregulated by agmatine and results in conversion of agmatine to putrescine, which can be subsequently converted into other polyamines or shunted into the TCA cycle for energy production. We have discovered an alternate agmatine operon in the Pseudomonas aeruginosa strain PA14 named agu2ABCA' that contains two genes for agmatine deiminases (agu2A and agu2A'). Agu2A' contains a twin-arginine translocation signal at its N-terminus and site directed mutagenesis and cell fractionation experiments confirmed this protein is secreted to the periplasm. Analysis of the agu2ABCA' promoter demonstrates agmatine induces expression of the operon during the stationary phase of growth and agu2ABCA' provides only weak complementation of aguBA, which is induced during log phase. We hypothesized the preferred expression during stationary phase, and the periplasmic location of Agu2A' may implicate agu2ABCA' in biofilm production. Biofilm assays of mutants of all three agmatine deiminase genes in PA14 revealed agu2ABCA', specifically its secreted product Agu2A', enhance the biofilm production of PA14 in the presence of agmatine. Without agu2ABCA', the PA14 biofilm is reduced in the presence of agmatine, showing agu2ABCA' reverses the biofilm response to exogenous agmatine and implicates role for polyamines in the biofilm development of P. aeruginosa.

We have previously shown agmatine to be present in cystic fibrosis (CF) sputum, positively correlated with inflammatory markers, and capable of inducing neutrophilic inflammation when injected into the lungs of mice. The acquisition of agu2ABCA' would potentially protect a pseudomonad in an inflammatory millieu that contains agmatine. Together these data reveal the dynamic role of agmatine and polyamine metabolism in the hostpathogen interaction. Background: Staphylococcus aureus chronically infects the airways of cystic fibrosis (CF) patients. To accomplish colonization and persistence, S. aureus is equipped with various proteins which facilitate adhesion and persistence. Limited knowledge is known about adaptive processes that occur during persistence in the host.

Aims: To assess the degree of adaptation, 77 S. aureus isolates of two patients (6.9 and 26.1 years old), who were chronically infected by S. aureus for 4.3 and 11.8 years, were further investigated.

Methods: Single or multiplex PCR was performed for the following adhesins: fnbA, fnbB, clfA, clfB, cna sdrD, sdrC, sdrE. Molecular typing of isolates was performed by pulsed-field gel electrophoresis (PFGE) and determination of agr-specificity groups. Biofilm formation was assessed by a microtitre biofilm assay.

Results: PFGE allowed to distinguish six and three different S. aureus strains in the first and second patient, respectively, with subtypes due to fragment pattern differences. The strains of the first patient belonged to agr group I (4 strains), III and IV (1 strain each) and the strains of the second patient to agr group I, II and IV. By PCR, all strains were positive for fnbA, fnbB, clfA, clfB, sdrC and 62/77 positive for cna. There were different sizes of amplicons in some strains and their subtypes for cna, fnbA, fnbB, clfA and clfB. Seventeen strains were negative for sdrD and 6 for sdrE.

Conclusions: By PCR analysis, strains were identified, which showed differences in amplicon sizes of special adhesins indicating changes in the repeat area assumingly due to adaptation of the microorganism to the hostile environment during longterm persistence. Cystic fibrosis (CF) airway infections are frequently polymicrobial. In addition, fastidious bacteria including anaerobes may be present yet undetected using standard microbial culture. Anaerobic bacteria may contribute to lung infections, however strict anaerobic culture is labor-intensive and inconsistent. To fully understand the complex CF airway microbiome, molecular detection techniques are needed that reliably quantify airway microbes. Quantitative real-time PCR (qPCR) offers rapid quantification of bacteria without culture. The reliability and validity of qPCR applied to CF airway specimens has not been determined.

Objective: We sought to determine the reliability and validity of qPCR microbial detection applied to CF airway specimens.

Methods: We obtained 152 airway specimens (throat swab, sputum, and saliva) from 16 CF subjects at three time points. We measured bacterial DNA copies for total bacteria, three typical CF pathogens and five anaerobic bacteria using qPCR. We determined sample processing variability by analyzing split samples from airway specimens. Each sample was analyzed in triplicate to determine reliability of our assays. We compared qPCR results to microbial quantitative culture for P. aeruginosa and S. aureus to determine validity. All values were log transformed.

Results: We performed 2,493 qPCR assays on 152 CF airway specimens. Considering samples with mean triplicate measurement ≥10 3 DNA copies, 96.1% had CV < 10%. For samples with 10 2 -10 3 DNA copies, 71.3% had CV < 10%. Split samples from the same specimens were highly correlated (r 2 =0.91, p<0.01) and of similar value (mean difference -0.02, 95% CI -0.09-0.04). Mean DNA copy number for P. aeruginosa and S. aureus correlated with quantitative microbial culture for specimen with ≥10 3 DNA copies (P. aeruginosa, r 2 = 0.55, p <0.01;S. aureus, r 2 = 0.73, p <0.01).

Conclusions: QPCR can reliably quantify aerobic and anaerobic bacteria present in DNA copies ≥ 10 3 from CF airway specimens. Application of qPCR to CF airway samples may improve our understanding of the CF airway microbiome.

This research was supported by grants from the Cystic Fibrosis Foundation (#ZEMANI08A0) and the National Institutes of Health (U01 HL081335). Background: Nontuberculous mycobacteria (NTM) are common in cystic fibrosis (CF) (13%) and prevalence increases significantly with age (1) . Current treatment regimens are complex and, particularly for M. abscessus in the setting of CF, may be ineffective. The use of amikacin is recommended for patients with rapidly growing NTM, extensive cavitary disease, and failed conventional treatments, but systemic use is associated with increased nephrotoxicity (15%), ototoxicity (37%), and vestibular toxicity (9%) (2) . Aerosolized amikacin has the potential for reduced toxicity. In this study we analyzed the effectiveness and toxicity of inhaled amikacin in patients with refractory NTM lung disease.

Methods: Records from an IRB-approved mycobacterial natural history study were queried to identify patients with refractory NTM lung disease who were treated with aerosolized amikacin and had a baseline and at least one follow-up visit.

Results: Between March 2003 through March 2009, 12 patients were identified with persistent M. abscessus (Mab) (n=9) and M. avium complex (MAC) (n=3) infections. Four (33%) patients had negative sputum cultures and 75% had a decrease in NTM culture growth quantity (e.g. heavy to light growth) by their final follow-up visit. Of the people who converted to negative sputum by their final visit, 3 of 4 had done so by 161 days. Fifty percent of patients showed improvement on serial CT scans. Ototoxicity was seen in two patients (17%) and resolved in one patient after returning to daily dosing instead of b.i.d.

Conclusions: Inhaled amikacin appears effective for treating refractory MAC and M. abscessus pulmonary infections with minimal toxicity. The observed time to improvement and sputum conversion suggest a minimum follow-up time of 6 months for future clinical trials. Background: Cystic fibrosis (CF) patients are prone to respiratory infections and biofilm formation in the lungs. Systemically-administered antibiotics are inefficient and lead to resistant species. Nanoemulsions (NE) are oil-in-water emulsions stabilized by surfactants with an average droplet size of ~400 nm. They have broad-spectrum antimicrobial activity and kill pathogens by interacting with their membranes. This physical kill-on-contact mechanism significantly reduces the possibility of the emergence of resistant strains. The NE is formulated from pharmaceutically acceptable ingredients. In this study we evaluated the minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of the nanoemulsion NB-400 against four genera of bacteria commonly infecting CF patients: Pseudomonas, Burkholderia, Acinetobacter and Stenotrophomonas. We also tested for potential synergy between NB-400 and colistin or tobramycin.

Methods: Multidrug-resistant gram-negative bacteria isolated from CF patients were tested in this study. These organisms include Pseudomonas, Burkholderia, Acinetobacter and Stenotrophomonas. The P. aeruginosa isolates had defined lipid A modifications that have been documented in many CF patients. MICs and MBCs were determined using CLSI guidelines and standard methods M7-A7 and M100-S17. The addition of alamar blue, a redox indicator that yields a colorimetric change in response to metabolic activity, was used to determine the MICs of NB-400 because of NE opacity at higher concentrations. Synergy between NB-400 and other antimicrobials was assessed using checkerboard synergy microtiter-based assays. The microbicidal activity of NB-400 has been visualized using transmission electron microscopy. values of ≥32/>32, ≥32/>32, 16/16 and >32/>32 µg/ml, respectively, against all strains. NB-400 was synergistic in vitro with colistin and not antagonistic to tobramycin. Transmission electron microscopy verified that NB-400 rapidly and effectively disrupts these CF pathogens.

Conclusions: NB-400 was effective against strains that were multidrugresistant, including colistin-resistant isolates of Burkholderia and Stenotrophomonas. None of the described lipid A modifications in Pseudomonas species impacted the MIC/MBCs with NB-400. Further stud-ies are ongoing to investigate the nebulization of NB-400 for the treatment of pulmonary infection in cystic fibrosis patients. Objective: The objective of this study is to determine the efficacy of the treatment protocol in eradicating PsA from patients with CF and to evaluate its effect on health related outcomes.

Methods: Electronic medical records of 257 CF patients who received treatment between 2002 and 2007 were reviewed for information regarding demographic, microbiologic, and therapeutic data. Successful eradication (SE) of PsA was defined as 3 consecutive cultures that had no growth of PsA within 6 months of the eradication attempt and while off antibiotic therapy. Subsequent to treatments targeted at PsA each attempt was evaluated for SE, duration of therapy, demographic data, body-mass index (BMI), and pulmonary function tests.

Results: Forty-one patients were identified who had their first PsA positive culture and underwent an eradication attempt. Of these patients, 34 (82.9%) experienced SE and 15 of the 34 (44%) subsequently had another PsA positive culture. Fourteen of these underwent a second eradication attempt with a success rate of 42.8%. Protocol medications were utilized in 68% of SEs and 32% of attempts were successful after the addition of inhaled colistin or IV antibiotics. Of all patients with SEs, 47% utilized inhaled medications alone. The patients who were SE were younger (46.9 ± 8.9 vs. 93.6 ± 25.8, p=0.046). There was no difference at initial PsA detection in percent-predicted lung function or BMI percentile between the patients that were eradicated and those who failed eradication (FEV 1 %: 92.0 ± 5.2 vs. 84.6 ± 3.7, p = 0.27), (BMI: 54.8 + 27.2 vs. 61.9 + 13.0, p = 0.82). Patients who were SE had a higher BMI at study conclusion compared to study enrollment (54.8 + 27.2 vs 63.7 + 23.1; p = 0.04). Twenty-three patients were treated based on the above protocol, 8 patients had inhaled colistin added to the regimen, and 8 required IV antibiotics.

Conclusions: Eradication of PsA can be achieved using a standardized protocol including inhaled and oral antibiotics. There was no difference in lung function or growth parameters between the two cohorts. However, the SE patients had a significant improvement in BMI at the study endpoint. 

MRSA) strains has increased. MRSA has been associated with increased hospitalizations, increased antibiotic use, and decreased lung function. It remains unknown if treatment of MRSA will improve outcomes. In response to a study by Ren et al. a novel eradication protocol was established: TMP/SMX for 4 weeks, plus mupirocin and rifampin during the last week of therapy. Patients who failed initial treatment or were allergic received linezolid for 2 weeks in place of TMP/SMX. Environmental contamination was addressed by replacing respiratory equipment, sanitizing surfaces with bleach, and treating close contacts with nasal mupirocin and antibacterial baths. The objective of this study was to determine the effectiveness of the eradication protocol for MRSA.

Methods: In this retrospective, observational study, data was collected from 10/1/2007-4/30/2008 from the Cystic Fibrosis (CF) Foundation Patient Registry, and electronic medical records from Intermountain Pediatric CF Center at Primary Children's Medical Center and the University of Utah. Inclusion criteria for this study was diagnosis of CF, first MRSA positive culture within past year, positive MRSA culture during data collection period, treatment with eradication protocol, and at least one follow-up culture. The primary outcome of this study was reduction or elimination of MRSA on culture. Other data collected were CF culture results, pre and post-treatment, oxygen saturations, antibiotics used, weight, and demographic data. A Mann-Whitney test and paired t-test were used for statistical analysis. This project was IRB approved.

Results: Thirty-six patient encounters were evaluated; 18 unique encounters met inclusion criteria (17 unique patients). The remaining patients had chronic MRSA for >1 year and did not meet inclusion criteria for analysis. MRSA burden was significantly decreased after treatment. The median MRSA burden before treatment was 1 vs. median after treatment of 0 (p=0.027). Ten of 17 patient encounters had complete eradication of MRSA at follow-up culture; only 1 patient has had recurrence of MRSA on subsequent follow-up culture. Sixteen of 17 received TMP/SMX as their antimicrobial agent; 1/17 received linezolid. The patient who received linezolid had previously failed the TMP/SMX regimen. Weight for patients between cultures was not significantly different (p=0.132). Oxygen saturation did not change with treatment (p=0.055).

Discussion: This study shows that successful eradication may occur by aggressively treating patients with MRSA. This population included patients found to be recently positive for this organism. The approach in our center addresses environmental contamination, nasal and skin carriage and treatment of contacts. Our approach utilizes an inexpensive oral regimen at first colonization. Linezolid use is carefully monitored. More studies are needed to see which treatments are most effective and to look for trends of increased antibiotic resistance and increased P. aeruginosa burden. Eliminating MRSA may be beneficial since previous retrospective studies have shown that chronic colonization with MRSA may lead to decreased lung function.

Faria, M.M. 1 ; Wong, J. 1 ; Rabin, H.R. 2, 3 ; Storey, D.G. 1, 2 1. Biological Sciences, University of Calgary, Calgary, AB, Canada; 2. Microbiology and Infectious Diseases, University of Calgary, Calgary, AB, Canada; 3. Medicine, University of Calgary, Calgary, AB, Canada P. aeruginosa strains adapt to the lung environment within the cystic fibrosis (CF) patients. Our hypothesis is that an alternate CFTR genotype, specifically the M1101K mutation, in CF patients may influence the P. aeruginosa lineages colonizing the lungs. The M1101K CFTR mutation is rare in the global CF population with an occurrence of 0.8%. However, founder effects and subsequent genetic isolation in the Hutterite brethren of North America has resulted in a predominance of this mutation in these CF patients. In order to test our hypothesis, we obtained isolates from two populations of patients that attended the UCMC Adult CF Clinic; M1101K CF patients and general CF patients. A total of 24 isolates from five M1101K patients were assessed and compared to 24 isolates from twelve general CF patients. Multi-locus sequence typing (MLST) revealed that in the 48 isolates there were 18 sequence types (STs) representing different P. aeruginosa lineages. Based on the P. aeruginosa MLST website (http://pubmlst.org/paeruginosa/), nine of those STs were previously identified in other CF isolates from Canada, the UK, the Netherlands, and Spain. In both patient populations, we found CF isolates that accounted for seven novel STs. Only two of our isolates had STs that were previously identified in a non-CF nosocomial isolate or an environmental isolate. In the isolates from the general CF population, three clonal complexes were present with the founder isolate for clonal complex 3, ST-17, previously identified in CF isolates from Canada and the UK with some ST-17 isolates reported as epidemic strains. There were also two clonal types, present in five of the general CF patients and in other CF isolates from Canada. Overall, these results suggested that in the general CF patients there was a predilection for P. aeruginosa lineages that were also identified in other CF patients rather than lineages found in environmental or nosocomial isolates. With respect to the P. aeruginosa isolates from M1101K patients, these patients were colonized with different lineages not seen in our general CF patients despite living in the same geographical region and attending the same CF clinic. Further, the trend was for the maintenance of one P. aeruginosa lineage in each M1101K patient. In comparison, isolates from our general CF population showed genetic variability within patients and overlap of STs between patients with lineages previously seen in CF isolates being favoured. Interestingly, the one exception was the M1101K homozygous patients that shared the same lineage, ST-689, suggesting a preference for this lineage in these patients. Overall, the isolates from M1101K patients revealed a more genetically homogenous and stable population of P. aeruginosa in each patient. Taken together, these results suggested that the M1101K CFTR allele might play a role in the predilection for or maintenance of P. aeruginosa lineages in these CF patients.

This study was supported by a grant from the Canadian Cystic Fibrosis Foundation to D.G.S. (Parkins et al. Pediatr Pulmonol 2008; 43:490-7; Sibley et al. PNAS 2008; 105:15070-15075) . Our lab has previously shown that about 40% of pulmonary exacerbations may be due to SMG in our adult CF population (Sibley et al. PNAS 2008 105:15070-15075) .

Several collections of the SMG have been phenotypically characterized in attempts to determine the core traits shared by these pathogens, and to gain a better understanding of possible virulence mechanisms. However, very few CF isolates have been characterized. Our novel collection consists of strains isolated from CF sputum by McKay Agar culture during periods of clinical stability and pulmonary exacerbation from 50 adult patients. The 129 CF isolates include 22 from patients where the SMG was associated with pulmonary exacerbations, while 107 of the isolates were present at lower concentrations in sputum and not directly correlated with exacerbations. We included an additional 11 strains of SMG isolated from invasive diseases (pus or blood) for comparison, as well as three reference strains.

We assessed all strains for hemolysis on blood agar, Lancefield grouping, production of hyaluronidase, chondroitin sulfatase, DNase, protease, hydrogen peroxide, intermedilysin, and performed biochemical profiling with API strips. In a subset of strains, we used a Vibrio harveyi bioassay to detect the production of the quorum sensing signal molecule AI-2, which has implications in polymicrobial infections as AI-2 has been found to upregulate the expression of virulence factors in another CF pathogen, Pseudomonas aeruginosa (Duan et al. Mol Microbiol 2003; 50:1477 -1491 . In brief, we found that all S. intermedius were devoid of hemolysis and Lancefield groups, but were uniform in the production of hyaluronidase, chondroitin sulfatase, DNase, and protease. S. constellatus strains were predominantly beta-hemolytic, Lancefield-group C which possessed the four hydrolytic enzymes assayed. Interestingly, this was the only S. constellatus biotype associated with pulmonary exacerbation in three patients. S. anginosus fell into a wide range of phenotypes. None produced chondroitin sulfatase, and only beta-hemolytic, Lancefield-group C strains produced hyaluronidase. Intermedilysin detection by PCR was exclusive to S. intermedius.

AI-2 production was assayed from supernatants collected after 24 hrs of growth in Todd-Hewitt-Yeast broth. All strains showed levels ranging from 5 to 552 times greater than the negative control. Although great variability was observed within each species, on average, AI-2 detection was highest in S. anginosus, followed by S. constellatus, and then S. intermedius.

To our knowledge, this is the first study to characterize SMG isolates from cystic fibrosis patients.

The Pseudomonas aeruginosa (Pa) type III secretion system (TTSS) and associated toxins make up an important virulence mechanism for Pa, causing death of epithelial cells, neutrophils and macrophages by direct membrane damage, injection of exotoxins and induction of airway inflammation. KB001, a PEGylated, recombinant human, Fab' antibody fragment is directed against PcrV, a structural component of the Pa TTSS. KB001 incapacitates the TTSS in all pathogenic Pa strains tested. A healthy adult volunteer study of KB001, up to 10 mg/kg, was tolerated and non-immunogenic.

Methods: This multicenter, randomized, double-blind, placebo-controlled safety and pharmacodynamic study was conducted at centers of the Cystic Fibrosis (CF) Foundation's Therapeutic Development Network. Two cohorts of 12 patients were randomized (2:1) to receive a single IV infusion of KB001 or placebo at 3.0 mg/kg or 10 mg/kg. Follow-up was 8 weeks. Patients had a sputum Pa burden of ≥ 1x10 5 CFU/gm, were on stable CF therapy and not on inhaled antibiotics from Days 0 through 28. Primary objectives: safety and tolerability. Secondary objectives: evaluate pharmacokinetics (serum and sputum) and potential immunogenicity. Exploratory outcome measures included changes from baseline in: sputum total and PcrV-expressing Pa burden by PCR and RT-PCR; microbial ecology (burden, predominance and diversity) by culture-independent genomic analysis; sputum and serum inflammatory markers; lung function and pulmonary symptoms.

Results: Preliminary study results suggest a single dose of KB001 at dose up to 10 mg/kg was safe and well tolerated. No patient developed anti-KB001 antibodies. Preliminary data revealed high baseline microbial burden of Pa and all subjects with evaluable sputum samples were positive for PcrV expression, and the presence of exotoxin genes. Subjects had large numbers of bacterial taxa despite baseline CF antimicrobial therapy. Although not statistically significant, a trend to reduction in sputum mucoid Pa burden was seen at the highest dose. Evidence for a potential anti-inflammatory, protective effect of KB001 was suggested by the changes noted in several sputum biomarkers.

Conclusions: Data from this study suggest that KB001 may have a bioprotective, anti-inflammatory impact in the CF airway. Larger randomized studies with multiple doses will be required to confirm the initial antiinflammatory effects observed in this pilot study and determine if Pa burden is reduced. Background: Patients with cystic fibrosis (CF) who have identical CFTR genotypes can have marked variation in pulmonary phenotype. Some of this variability has been explained by specific environmental factors; however a portion remains unexplained and may be due to non-CFTR genetic factors. Innate immune inflammatory responses might play a role in the pathogenesis of airway disease in CF. Specifically, inter-individual variation in Toll-like receptor (TLR)-mediated innate immune responses might account for a portion of the observed variation in pulmonary phenotype.

Hypothesis: We hypothesize that TLR-mediated innate immune inflammatory responses measured in the whole blood of CF patients, ex vivo, is associated with lung function decline.

Methods: We enrolled CF patients in whom at least 3 weeks had passed since their last acute pulmonary exacerbation and who were not taking immunosuppressive medications. We cultured citrate-anti-coagulated whole blood obtained from each subject with a panel of agonists for specific TLRs (E.coli 0111:B4, P.aeruginosa penta-acyl and hexa-acyl LPS:TLR4, flagellin:TLR5, pam3CSK4:TLR2/1, FSL-1: TLR2/6, and CLO97: TLR7/8) and measured the production of inflammatory mediators (IL-6, IL-8, MCP-1, TNFα, and IL1ra) by immunoassay. Peripheral monocyte counts were obtained for each blood sample. Associations between log 10 -transformed, monocyte-count normalized TLR agonist-induced mediator production and mean FEV1% was tested in a GEE regression model with robust standard errors using data from the 5 years prior to sampling available in PORT CF.

Results: We enrolled and analyzed 65 subjects who had a mean FEV1% of 73.6% (StdDev 22.1). We identified significant (p<0.05) associations between higher flagellin-induced IL-6, IL-8, TNFα, and IL-1ra and higher mean FEV1%. Similar relationships were identified between hexa-acyl LPS induced IL-6, IL-8, and IL-1ra responses and FEV1%. In whole blood incubated with media alone, only levels of IL-6 showed a significant relationship with FEV1%.

Conclusions: Innate immune responses to a subset of TLR agonists, measured in whole blood ex vivo, are inversely associated with lung function in CF patients. These findings suggest that preservation of innate immune responses could have beneficial effects on lung function in CF. 1 1. Pediatric Pulmonology, Ghent University Hospital, Ghent, Belgium; 2. Research Laboratory Microbiology, Ghent University Hospital, Ghent, Belgium Introduction: Longitudinal data on genotypes of Pseudomonas aeruginosa strains after eradication treatment are limited. We followed cystic fibrosis (CF) patients after a first-ever P. aeruginosa isolate and evaluated the P. aeruginosa-free time period after eradication therapy and the efficacy of treatment, relying on time to subsequent isolation and on comparison of the genotypes of subsequent P. aeruginosa isolates.

Methods: Between January 2003 and December 2008 sputa or nasopharyngeal aspirates were cultured prospectively from 41 CF patients with a first-ever P. aeruginosa isolate. Eradication treatment consisted of oral ciprofloxacin and inhaled tobramycin or colistin for 3 months. Genotyping of the cultured P. aeruginosa isolate was carried out by RAPD-analysis.

Results: Eradication treatment failed in 7 patients, who became immediately colonized. The first eradication treatment was successful in 34 (83%) patients, of which 19 (56%) had a second P. aeruginosa infection after a median period of 12 months (range 5 -58 months), with 4 patients becom-ing chronically colonized after a median time of 13 months (range 5 -36 months). At present, 15 patients (37%) remain free of P. aeruginosa after a median follow-up period of 50 months (range 5 -63 months).

Sixteen of the 26 patients (62%) with at least a second isolate showed an identical genotype. Ten out of these 16 became chronically colonized. The 11 patients becoming chronically colonized during the study period had a significantly shorter P. aeruginosa-free interval between the first-ever and second isolate compared to not chronically colonized patients (median 3, IQR 3-9 months versus 15, 10-31 months; p = 0.041) and 10 of the 11 had the same genotype for first and follow-up isolates.

Conclusion: Our findings of eradication success and median time to recurrence after a first positive P. aeruginosa culture are comparable to those of three earlier studies. However, we report longer lasting eradication. In addition, our results indicate that the presence of an identical P. aeruginosa genotype in the second positive culture and a short P. aeruginosa-free interval are ominous signs of impending chronic colonization. Background: Progressive decline in lung function is a defining feature of cystic fibrosis (CF). Yet, clinical course varies considerably even among children carrying the same genetic mutations. It is unclear which children are at risk for earlier and more severe lung disease and who will experience a more rapid decline in lung function. Since airway inflammation plays a central role in CF lung disease, biomarkers of inflammation that can be used to monitor disease activity would be extremely valuable.

Objective: To examine cross-sectional and longitudinal relationships between sputum biomarkers of inflammation and lung function measures.

Methods: In this longitudinal, observational study, we measured airway cytology and a panel of proteases (free neutrophil elastase, MMP-9), antiproteases (elastase-alpha1 antiprotease complexes=NEAPC, SLPI, TIMP-1), and cytokines (TNF-α, IL-1β, IL-6, IL-8) in induced sputum collected from 35 school age children with CF (20M:15F; age 11±3 years; FEV1 95±15% predicted) on an annual basis over 3 years during times of clinical stability. Serial pulmonary function testing (FVC, FEV1, was performed at the time of sputum induction. Statistical analyses were performed to answer these questions: (1) Do sputum biomarkers of inflammation change over time? (2) Do changes in sputum biomarkers relate to changes in lung function? (3) Can sputum biomarkers predict subsequent changes in lung function?

Results: Sputum induction yielded an adequate sample for analysis approximately 80% of the time. The rate of decline in FEV1 in our cohort was -1.1 %predicted/year. Relatively small but statistically significant increases in sputum elastase (p=0.01), TIMP-1 (p<0.01), and TNF-α (p=0.02) and a near significant decline in SLPI (p=0.08) were observed over time. Changes in FEV1 correlated with changes in % neutrophils (r=-0.38, p=0.02), TIMP-1 (r=-0.37, p=0.03) and weakly correlated with changes in IL-8 (r=-0.29, p=0.09) and NEAPC (r=0.29, p=0.09). To examine the predictive ability of sputum biomarkers, we first categorized subjects as either lung function "decliners" or "maintainers" by comparing subject specific slopes for FEV1 to CF-specific reference equations (Kulich M et al. Am J Respir Crit Care Med 2005; 172:885) . Using this approach, 19 subjects were "decliners" and 16 were "maintainers." A logistic regression model revealed that the initial measurement of TIMP-1 had the highest individual predictive value for subsequent lung function decline (c=0.78) while TIMP-1, SLPI and TNF-α had the highest combined predictive value (c=0.77). Examining the predictive value of biomarker slopes (i.e changes over time), the single best predictor of lung function decline was NEAPC (c=0.67) while the best predictive combination was elastase, NEAPC and TNF-α (c=0.79).

Conclusions: This is the first comprehensive longitudinal assessment of the relationship between airway inflammation and lung function in school age CF children. Changes in individual sputum biomarkers do relate to changes in lung function. Single and serial determinations of sputum biomarkers of inflammation have predictive value for lung function decline.

Supported by CF Foundation (#SAGEL07A0), K23 RR018611-05 and Colorado CTSA #1 UL 1RR014780 (NCRR/NIH), and U01 HL081335 (NHLBI/NIH).

Kerem, E. 1 ; Wilschanski, M. 1 ; Ajayi, T. 2 ; Miller, N. 2 ; Pugatsch, T. 1 ; Armoni, S. 1 ; Cohen, T. 1 ; Reha, A. 2 ; Constantine, S. 2 ; Elfring, G. 2 ; Miller, L. 2 1. Hadassah Hebrew University Hospital, Jerusalem, Israel; 2. PTC Therapeutics, South Plainfield, NJ, USA Background: Though chronic coughing is among the most prominent of cystic fibrosis (CF)-related symptoms, objective cough assessment has seldom been performed in patients with CF. The LifeShirt ® (VivoMetrics, Ventura, CA) is a new device that quantitatively measures cough. The LifeShirt ® incorporates motion-sensing transducers, electrodes, a microphone, and a 3-axis accelerometer into a lightweight, washable vest. The frequency and intensity of cough can be measured using integrated input from the motion sensors and microphone. The device is designed to discard events such as throat-clearing, sneezing, sighing, or talking. Data are stored on a compact flash card housed within the recorder and can be uploaded for analysis using specialized software. We used the LifeShirt ® to characterize coughing in patients with CF.

Methods: As a prelude to a Phase 2 study of ataluren (PTC124™), subjects were fitted with the LifeShirt ® in the clinic and were instructed to wear the vest at home or in the community for 24 hours before returning it by mail to the clinic. Data recordings were transmitted from the clinic to VivoMetrics for assessment within 24 to 48 hours. Waking and sleeping periods were derived from LifeShirt ® data regarding body position and breathing. In addition, FEV 1 [1] [2] [3] for daytime and 1 [1] [2] [3] for nighttime coughing. Cough frequency tended to increase with lower FEV 1 and greater age; no obvious gender effect was seen. Some patients reported occasional neck discomfort associated with the throat microphone; no other device-related adverse events were reported.

Conclusions: Patients with CF cough frequently, especially while awake. While subjective evaluation may provide useful information, objective evaluation during wake and sleep states may be important to fully quantify the impact of coughing and may offer a clinically meaningful outcome measure of how patients feel and function for use in CF therapeutic trials. In conjunction with patient-reported subjective assessment, the LifeShirt ® will be used for objective assessment of coughing in a Phase 3 study of ataluren (PTC124™) as a treatment for nonsense mutation CF. 

We used the CF Foundation Patient Registry to assess variation of the pulmonary exacerbation rate with the influenza season from July 2003 through June 2007. The outcome of interest, monthly CF pulmonary exacerbations, was defined as treatment of a respiratory illness with IV antibiotics. Duplicate exacerbations in a calendar month were excluded. Influenza season was defined as all months during which the Centers for Disease Control national surveillance data for influenza exceeded epidemic thresholds. The unit of analysis was the exacerbation. We calculated incidence rates per 10,000 person-months, incidence rate ratio seasonal comparisons, and estimated excess events.

Results: In 2003, the cohort had 21,506 patients. Of these, 8625 (40.1%) were adults. The majority of adults and children were white (92.3% and 86.13%) and male (53.1% and 51.3%). Adults and children had a median of four outpatient clinic visits (IQR 1-7 and 2-6). On average, adults had worse lung function than children (FEV1 % predicted; 60.8% vs. 86.8%) and lower percent height-weight measures above US average (5.0% vs. 44.8%). From the 2003 through 2007, there was an annual increase in the incidence rate of exacerbations that was temporally associated with influenza activity (Figure) . Exacerbation incidence rates peaked every January for adults and children. Overall influenza season incidence rates were 873.1 per 10,000 person-months among adults and 452.8 per 10,000 person-months among children. Seasonal incidence rate ratios were 1.13 (95%CI: 1.11, 1.16) among adults and 1.13 (95% CI: 1.10, 1.16) among children. The average influenza season during the study was four months per year. Exacerbations during these periods contributed to an excess of pulmonary exacerbations by 99.3 per 10,000 person-months among adults and 53.4 per 10,000 person-months among children. Assuming a four month influenza season annually, we estimate that there are 397 exacerbations per 10,000 adults and 214 exacerbations per 10,000 children that are attributable to influenza every year.

Conclusion: Our data demonstrate an increase in exacerbations that is coincident with influenza activity. Furthermore, we estimate that influenza contributes to a substantial annual excess of exacerbations. Studies to determine any causal link between influenza and CF morbidity are warranted.

Supported: NHLBI (HL72017, HL007287). Recent studies have shown that exposure to ambient air pollution is associated with reduced levels of lung function in healthy children. Because children with cystic fibrosis (CF) have impairment in lung function, they may be highly susceptible to effects of air pollution on lung function.

Methods: Data on children with CF were obtained from the CFF Patient Registry. Ambient air pollutant concentrations were obtained from the US Environmental Protection Agency. Study inclusion criteria were: age 6 to 16.5 years and at least one spirometry from 1994 to 2006. The cohort was split into 50% development cohort and a 50% validation cohort. Exposure to ambient fine particulate matter (PM 2.5 ) was estimated using the year 2000 average concentration at the nearest monitor within 30 miles of the home zip code centroid. The cross-sectional impact of PM 2.5 on FVC and FEV 1 at age 6 years was assessed using gender-specific longitudinal mixed effects regression models with random slopes and intercepts. The models adjusted for age, year of birth, and age-adjusted height residuals employing linear splines.

Results: Our development cohort (n=2583 males and 2413 females) mirrored the overall eligible National CF pediatric population with a baseline age of 9 years, with an average FEV 1 % of predicted of 87.5% and 84.8% in males and females respectively.

Conclusions: Long-term exposure to ambient PM 2.5 is associated with decreased level of lung function in children with cystic fibrosis. Longitudinal analyses assessing the reduction in growth of FEV 1 and FVC will be presented.

Supported by EPA, the Cystic Fibrosis Foundation Therapeutics, NHLBI (K23 HL72017-01), NIEHS (R21 ES015841-01), NIEHS (K24ES013195).

Adjusted for age-adjusted height, and year of birth Insurance status (SES) did not influence the point estimates or 95% CI Patients (pts) with cystic fibrosis (CF) suffer frequent pulmonary exacerbations, diagnosed by decline in lung function and increased symptoms. Early and aggressive therapy is associated with better outcomes. CF pts do not usually monitor their lung function or symptoms at home, and exacerbations are only detected when pts are sick enough to seek care. The purpose of this study was to assess the feasibility of home monitoring using the AM2 electronic spirometer to measure daily FEV1 and symptoms, to determine our subjects' ability to transmit data using telephone modem, and to determine if exacerbations can be predicted.

Methods: This is a six month study of ten pts with CF. Each was given an electronic device to measure FEV1 twice daily and symptoms once daily. Data was transmitted to us weekly by phone, and exacerbations were scored using the Akron exacerbation scale. Medical records were examined for evidence of clinically detected exacerbations.

Results: The mean age was 22.5 ± 8.7; 50% were female, and the median FEV1 was 78.5% predicted. Pts completed home monitoring over 85.7% of days and were able to transmit their data through telephone modem weekly. There were 28 exacerbations detected through home monitoring, but only 8 of these were clinically detected and reported by the pts to their physicians. Sixteen exacerbations were detected by FEV1 alone, and 12 by FEV1 and symptoms combined. Home monitoring detected exacerbations an average of 15.6 days before pts contacted the care center because of symptoms. In the six months prior to the study, the number of IV antibiotic courses was 0.71±0.28 which fell to 0.14±0.14 during the study period. Likewise, the number of oral antibiotic courses before the study was 1.28±0.42 compared with 0.86±0.40 during the study period. The average total antibiotic courses before the protocol were 2.0 and during the protocol was 1.0 (p=0.11). Finally, mean FEV1 was 2.76 during the 6 months prior to the study, and was 3.06 during the study (78.5% and 83.0% predicted, respectively, p=0.10).

Home spirometry and symptom monitoring is feasible as evidenced by a high rate of participant adherence (85.7% of days) and successful transmission of the data. Although these are pilot study results, the data suggests home monitoring can detect exacerbations more than two weeks before patients typically seek medical care; therefore, early intervention may improve CF outcomes. Although not statistically significant, there was less use of IV and oral antibiotics during the study period compared to the prior 6 months. Though speculative, this may be a result of objective feedback leading to better treatment adherence resulting in less need for antibiotics. There was also a clinically meaningful improvement of 5% in FEV1 during the home monitoring period, though it was not statistically significant in this pilot study. Therefore, home measurement of FEV1 could be used to promote earlier, more aggressive pulmonary therapy, to prevent more severe exacerbations, possibly improving long-term lung function. A randomized controlled trial is planned to further assess these findings. Background: Serum procalcitonin (PCT) values have been introduced to differentiate between septic and other infections in pediatric patients but have not been studied in children with cystic fibrosis (CF). Early recognition of pulmonary exacerbations is necessary for optimal management of CF patients and sensitive biomarkers to do so are lacking.

Aim of the Study: To establish baseline values for PCT in children with CF and to compare these to values at onset of a pulmonary exacerbation.

Methods: In all CF patients 1 to 19 years old, blood samples were drawn during outpatient clinic visit, and in case of a pulmonary exacerbation meeting Fuchs criteria if treatment with intravenous antibiotics occurred during the subsequent year. Serum PCT was determined using a quantitative immunoassay (BRAHMS Kryptor PCTsensitive, Henningsdorf, Germany). The study was approved by the hospital's Ethics Committee.

Results: Mean age of 92 subjects was 10 years 0 months (SD 4 years 8 months), mean FEV 1 89% (SD 18%). Nine patients (9.8%) were chronically colonized by Pseudomonas aeruginosa. Mean baseline PCT value was 0.06 ng/ml (sem 0.005 ng/ml), C-reactive protein (CRP) 2.3 mg/L (sem 0.5 mg/L), White Blood Cell count (WBC) 9808 x 106/L (sem 271 x 106/L). Mean PCT in 32 subjects subsequently admitted for treatment with IV antibiotics was 0.07 ng/ml (sem 0.005 ng/ml) (Paired T test with baseline values p = 0.075) at onset and 0.07 ng/ml (sem 0.005 ng/ml) at end of treatment. Mean age of these 32 subjects was 13 years 5 months (SD 5 years 0 months), mean FEV 1 on admission 55% (SD 17%), mean FEV 1 at discharge 64% (SD 16%). Two subjects had outlier values: one subject had catheter sepsis (PCT 0.43 ng/ml, CRP 33.3 mg/L) and one acute gastro-enteritis (PCT 0.48 ng/ml, CRP 14.7 mg/L).

Conclusion: Baseline PCT values in CF children are not different from values reported in healthy children. In CF children PCT values do not rise significantly at the onset of a respiratory exacerbation and thus hold no promise as an early marker to identify a pulmonary exacerbation. PCT values were markedly higher in cases of acute infections such as catheter sepsis and acute gastro-enteritis. Background: Inhaled corticosteroids (CS) for obstructive airway disease in cystic fibrosis (CF) have not been shown to improve overall lung function, and are not currently recommended. Less well studied is their role in decreasing the number of pulmonary exacerbations. We aim to study the efficacy of inhaled CS and corticosteroid/bronchodilator (CS/BD) combina-tion therapy in decreasing the frequency of exacerbations in CF lung disease.

Methodology: We retrospectively reviewed a cohort of patients over the last 14 years at an accredited academic CF care center using the Port CF© database. Included patients were using nebulized dornase alpha, a short-acting bronchodilator and an oscillatory compression device for airway clearance. All episodes of pulmonary exacerbations and utilization of inhaled CS (e.g., budesonide, triamcinolone, fluticasone, etc.) as well as CS/BD therapies (e.g., fluticasone/salmeterol, budesonide/formoterol) were identified. Using Chi-Square testing, the Mantel-Haenszel odds ratio was calculated to assess the likelihood of pulmonary exacerbations when using either CS or CS/BD combinations. The odds ratios were also calculated individually for 3 subgroups of disease severity as classified by %FEV 1 predicted, i.e., mild 70-89%; moderate 40-69%; severe <40%.

Results: We identified 11,201 pulmonary exacerbations from a total of 12,809 clinical care episodes, with a median patient age of 17±10 years and a male:female ratio of 0.7. The odds ratios of developing a pulmonary exacerbation when using inhaled CS and CS/BD combination therapy were 1.124 (C.I. 0.98-1.29, p=0.1) and 1.108 (C.I. 0.99-1.24, p=0.08) respectively. When grouped by disease severity, there was a reduction in pulmonary exacerbations that attained statistical significance only in the subgroup of patients with moderate CF lung disease using inhaled CS alone (OR 0.46, p<0.0001) .

Conclusions: In our study, inhaled CS and CS/BD combination therapy did not decrease the overall number of pulmonary exacerbations. Although there was a statistically significant decrease in exacerbations in the subgroup of patients with moderate disease using inhaled CS alone, the implications of this are uncertain; a prospective randomized controlled trial is needed to better delineate this association. We support the current recommendation against the routine use of inhaled corticosteroids in CF lung disease. Each year, nearly 40% of cystic fibrosis (CF) patients in the US are treated with IV antibiotics for a pulmonary exacerbation. Pulmonary exacerbations are associated with poorer quality of life, poorer survival, and faster subsequent FEV 1 decline. The impact of pulmonary exacerbations on subsequent FEV 1 decline has not been fully explored. We hypothesized that more frequent pulmonary exacerbations would be associated with faster FEV 1 decline in subsequent years in adults and children with CF.

Methods: Subjects were individuals in the CF Foundation Patient Registry who were ≥6 years old in 2003 and had ≥2 FEV 1 measurements/year in [2003] [2004] [2005] [2006] . To compare the rates of FEV 1 decline (% predicted/year) between patients with 0, 1, 2, and 3+ exacerbations treated with IV antibiotics in 2003, mixed effects modeling was used. We adjusted for the following characteristics known to impact FEV 1 : gender, socioeconomic status (Medicaid insurance), BMI, CF-related diabetes, baseline FEV 1 , and persistent colonization with P. aeruginosa, B. cepacia, and MRSA (defined as ≥2 positive respiratory cultures in 2003). Other potential confounders and interactions terms were included in a final model only if they added significantly to this baseline model.

A total of 8490 subjects met selection criteria: 24% were adults, 50% were male, 48% were homozygous ∆F508, and 41% were persistently infected with P. aeruginosa in 2003. Of these, 60% had 0 exacerbations in 2003, 23% had 1, 10% had 2, and 7% had 3 or more. In 2004, 63% of subjects had the same number of exacerbations, 21% had more, and 16% had fewer. The Table lists the mean (95% CI) additional rate of FEV 1 decline in 2004-2006 for subjects with pulmonary exacerbations in 2003. Pulmonary exacerbations had a significant impact on subsequent FEV 1 decline. This impact was larger in children than in adults.

Conclusions: FEV 1 decline is accelerated for children experiencing more frequent pulmonary exacerbations, and for adults treated for ≥3 exacerbations/year. Prevention of pulmonary exacerbations may lead to improvements in FEV 1 decline. Background: Cardiac diseases are rarely reported complications of cystic fibrosis (CF), with the exception of cor pulmonale. Cases of arrhythmia in CF patients have been attributed to underlying right-heart failure or arrhythmogenic medications, although research exploring the physiologic role of CF transmembrane conductance regulator (CFTR) in myocardial hypertrophy, cardiac ischemia, heart failure, and arrhythmogenesis is underway.

Objective: We report a series of eleven adult CF patients with documented supraventricular tachycardia (SVT), and investigate for associations between CF disease characteristics and SVT.

Methods: Referral records to Cardiology from the Adult Cystic Fibrosis clinic at St. Michael's Hospital in Toronto, Ontario, Canada from 2002 to 2009 were reviewed to identify patients with CF and confirmed SVT. Clinical information, including demographics, CF genetics, disease manifestations, treatment with potentially arrhythmogenic medications, history of cardiac diseases, anemia, or thyroid disease, and results of cardiac investigations were recorded.

Results: Eleven patients were identified with CF and SVT. SVT diagnosis was confirmed with electrocardiogram (ECG), Holter monitor, loop recorder, and/or electrophysiology (EP) studies. One patient was diagnosed with Wolfe-Parkinson-White (WPW) syndrome, six with atrioventricular nodal-reentrant tachycardia (AVNRT), one with atrioventricular reentrant tachycardia, and three with unspecified supraventricular tachycardia. CFTR genotypes included two ∆F508 homozygotes, five ∆F508 heterozygotes, three non-∆F508 heterozygotes, and one with unknown genetics. Lung function (FEV1) at time of diagnosis of SVT ranged from 23% to 75% predicted. No patients had a history of cardiac disease or clinical evidence of cor pulmonale. Four patients were investigated with echocardiograms, all of which showed normal left and right ventricular function and normal chamber dimensions. All patients had hemoglobin levels greater than 100 mmol/L and there were no identified thyroid hormone abnormalities. Nine of eleven patients were treated with β-agonists and no patients were treated with theophylline or atropine. For management, four patients underwent slow pathway ablation for AVNRT, one patient (with WPW) was treated with fle-cainide, three patients were treated with diltiazem, and three patients received no specific treatment.

Conclusion: Although arrhythmias are well-described in chronic lung diseases such as COPD, reported cases of arrhythmias in patients with CF are rare. We report the largest case series of SVT in adult patients with CF to date. No clear association of SVT with CFTR genotype, CF disease manifestations, or treatment was determined. In contrast to previously published case reports, these eleven patients did not have underlying cor pulmonale and were not treated with theophylline or atropine. A combination of clinical factors may predispose certain patients with CF to development of SVT, or there may, in fact, be no association present between the two conditions. Advances in molecular cardiology research regarding the physiologic role of CFTR in the heart may provide further clarification. Junge, S.; Brinkmann, F.; Ballmann, M. paediatric, medical school, hannover, Germany Background: Allergic bronchopulmonary aspergillosis (ABPA) is a rare complication of cystic fibrosis (CF). Sensitization to Aspergillus fumigatus leads to allergic inflammation of the lower respiratory tract. Persistent ABPA is suspected to cause permanent pulmonary damage. There are few data on long-term development of lung function in these patients.

In this study we want to evaluate whether ABPA episodes significantly alter the deterioration of lung function in patients wirh CF.

Methods: In a retrospective case-controlled study we collected data from 26 patients who were diagnosed with ABPA (Stevens criteria CID, 2003) and their matched pairs. Patients and controls were matched for age, gender, Pseudomonas infection status and lung function. They were followed-up for 1-12 years (mean 5.7y) after diagnosis of ABPA. All patients were treated with corticosteroids (initially 0.5-2mg/kg) and most of them with itraconazole.

Results: Sixteen of 26 patients (62%) experienced mostly 2 relapses of ABPA, the majority (69%) of them within a year after cessation of corticosteroids. Fifteen patients (9 girls, 6 boys, mean age 7.2 years, mean FEV1 91.9% before diagnosis, 80% with chronic Pseudomonas infection) were followed-up for more than 5 years. The decrease in FEV1 was not significantly different from the control group (ABPA 1.6% (SD 18.6) vs. no ABPA 1.6% (SD 12.0)). Even in the subgroup of 9 patients with 2 and more relapses there was no statistical difference.

Conclusion: The long-term outcome of lung function in our group of CF patients with ABPA is comparable to those without ABPA, maybe due to timely and aggressive therapy with systemic steroids. In the current study, immunohistochemical methods were used to investigate the biodistribution and cellular uptake of amikacin in rats following a single 90 mg/kg exposure to liposomal amikacin (Arikace™) and to free amikacin.

Methods: Healthy, adult female rats were exposed to either a single dose of aerosolized liposomal amikacin (Arikace™) or free amikacin by inhalation, both at a nominal dose of 90 mg amikacin per kg body weight. Rats were sacrificed immediately post dosing and at 1, 4 and 24 hours post dosing. The right lobes of each lung was homogenized and the total amikacin level in each supernatant measured using a commercial immunopolarization fluorometry method (Abbot TDx). The left lobes were frozen immediately and sectioned on a cryostat. Each section was fixed in a mixture of 0.2% glutaraldehyde / 2% paraformaldehyde and permeabilized in 0.1% Triton X-100. After permeabilization, the tissue sections were stained with rabbit anti-amikacin serum and goat anti-rabbit antibody conjugated to phycoerythrin and then examined microscopically.

Results: Exposure to liposomal amikacin (Arikace™) and to free amikacin alone produced diffuse distribution of amikacin throughout the lung within one hour after the end of inhalation. However, over the next 24 hours, this diffuse staining decreased slowly for the liposomal amikacin (Arikace™)-treated rats with a concurrent increase of brightly stained macrophages. In comparison, the staining intensity within the lung decreased rapidly following the exposure to free amikacin. More importantly, the free amikacin group exhibited fewer fluorescent macrophages at the later necropsy times (4 and 24 hours post dosing). No other cell types were observed to have amikacin levels above background in either treatment group. These results are in agreement with lung amikacin levels determined by TDx, which showed higher concentrations of amikacin and a slower pulmonary clearance of amikacin for the liposomal amikacin (Arikace™)-treated rats.

Conclusions: Immunofluorescent staining showed that the inhalation of either liposomal amikacin (Arikace™) or free amikacin results in a diffuse distribution of amikacin throughout the lung immediately after dosing. However, unlike the rapid pulmonary clearance for the free amikacin treated rats, the liposomal amikacin (Arikace™)-treated rats exhibited slower pulmonary clearance and a greater degree of uptake by alveolar macrophages at later necropsy times. 4 1. Behavioral Medicine and Clinical Psychology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA; 2. Anesthesiology, Wake Forest University School of Medicine, NC, USA; 3. Anesthesiology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA; 4. Radiology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA Computed tomography (CT) scans in infants and children with cystic fibrosis (CF) have revealed early and previously undetectable alterations in lung structure that precede changes in pulmonary function. Children who cannot lie still and perform the respiratory maneuvers for CT scanning require sedation or general anesthesia (GA) to ensure motionless images and the necessary lung volumes needed to produce high-quality images. GA is reported to be more time efficient than conscious sedation with less side effects and negligible failure rates. The ProSeal™ LMA (laryngeal mask airway, PLMA™) under GA allows for higher inspiratory pressures without tracheal intubation for assisted ventilation with a conduit to suction the stomach. We report our experience with CT scans performed on children using GA and the PLMA™.

Methods: Patients were recruited from a larger clinical trial in 2-6 year olds with CF. CT scans were planned at baseline and 18 months later. Fourteen children had one scan and seven of the 14 had two scans totaling 21 scans. Children were induced with inhaled sevoflurane alone or with nitrous oxide. IV administration of 2 mg/kg propofol bolus was used to increase depth prior to placing the PLMA™. Sevoflurane and additional propofol boluses were used to maintain GA for the duration of the CT scan. Before obtaining inspiratory images, ventilating breaths were given at pressures between 20-25 cm H 2 O. PLMA™ was disconnected from the circuit and elastic recoil of the lungs provided expiratory images. The PLMA™ was removed prior to post anesthesia care unit (PACU) arrival. Scans were evaluated for motion, atelectasis and ability to detect air trapping by a radiologist on a 9 point scale (excellent 8-9, good 6-7, fair 5-6 and poor <5). Recorded times included: anesthesia (anesthesia start time to PACU arrival), PACU (PACU arrival to PACU discharge) and total time (anesthesia start time to PACU discharge).

Results: Fourteen of 16 families approached participated. Mean ± SD age was 37.1 ± 11.1 and 58.6 ± 10.3 months at the two CT scans. Mean ± SD for anesthesia, PACU and total times were 26. 7 ± 5.5, 36.7 ± 11.4, and 63.4 ± 11 .3 minutes respectively. There were no complications, cancellations or admissions. No complications or concerns were reported at 24 hour followup phone calls. One family declined the 18 month scan due to the child awaking upset after the baseline scan. CT scan quality was evaluated as excellent in 11, good in 9 and fair in 1. No scans were assessed as poor.

Conclusions: Use of CT to better understand early lung disease in children with CF is increasing in clinical settings and research studies. Our results show GA can be safely and efficiently administered in a clinical research setting. We found good subject acceptance and high-quality CT scans using GA and the PLMA™. This method is a viable option when high-quality CT scans are needed in children with CF.

Supported Background: In cystic fibrosis (CF) the pulmonary involvement frequently leads to severe impairment and disability. Consequently, it is clinically relevant to validate a multi-factorial scoring system to simultaneously assess pulmonary function, anatomical involvement and their clinical consequences. In this context, numerous radiographic and clinical scoring systems have been used, including the Shwachman-Kulczycki Clinical Score (SK) and the Brasfield Radiological Score (BR). On the other hand, Celli et al. (N Engl J Med 2004; 350:1005-12) developed the 0-10 BODE index which includes measurements of body composition (body mass index: BMI), airflow obstruction (forced expiratory volume in one second, FEV 1 ), dyspnoea (Modified Medical Research Council dyspnoea scale) and exercise capacity (six-minute walking distance, 6MWD) for patients with chronic obstructive pulmonary disease (COPD). This scheme has been found to predict exacerbations, hospitalizations and mortality, being also sensitive to detect changes over time after interventions in COPD patients. To the best of the author's knowledge, however, no previous study has investigated whether such a multi-dimensional scoring system is also useful in CF adult patients.

Objective: To evaluate the association between the BODE score with clinical (SK and exacerbations per year, Ex/y), radiological (BR) and functional (oxygen desaturation during submaximal exercise (6MWD), ∆SpO 2 ) indexes of disease severity in CF patients.

Results: Forty CF patients (23 men) aged above 18 years old were assessed. The Kolmogorov-Smirnov test showed normal distributions for all variables, except age and ∆SpO 2 . After linearization, however, ln∆SpO 2 showed a Gaussian distribution. The baseline characteristics were (mean±SD): age 28.23 ± 12.05 y; BMI: 21.03 ± 3.22 kg/m 2 ; FEV 1 2.10 ± 1.04L and 59.94 ± 24.33 %pred; 6MWD 587. 2 ± 100.11 meters; SK 75.38 ± 20.11; BR 7.76 ± 5.304; BODE 2.25 ± 2.035; ; ∆lnSpO 2 1.196 ± 1.0; Ex/y 1.75 ± 1.446.

BODE index was strongly correlated with SK and BR (r = -0.907 p < 0.0001 and r = 0.821 p < 0.0001). When contrasted to SK and BR, The BODE index showed similar levels of association with ∆SpO 2 and Ex/y (see table below: Pearson, r and Spearman, rho).

Conclusions: The BODE index satisfactorily reflects the clinical status of CF adult patients assessing multiple aspects of the pulmonary-related disfunction. This is a simple multidimensional grading system, that needs a less complex clinical analysis and no radiological exams. So, it might prove to constitute a practical tool for adult patients follow-up.

p < 0.001

Loeve, M. 1,2 ; Gerbrands, K. 1 ; Hop, W.C. 3 ; Rosenfeld, M. 4 ; Tiddens, H.A. 1, 2 1. pediatric pulmonology, Erasmus MC Sophia, Rotterdam, Netherlands; 2. radiology, Erasmus MC, Rotterdam, Netherlands; 3. biostatistics&epidemiology, Erasmus MC, Rotterdam, Netherlands; 4. pediatric pulmonology, University of Washington School of Medicine, Seattle, WA, USA Background CT is more sensitive than pulmonary function tests (PFTs) to track progression of cystic fibrosis (CF) lung disease. Whether CT parameters are predictive for respiratory tract exacerbations rate (RTE-R) in an unselected cohort is unknown.

Aim: To establish the predictive value of CT regarding RTE-R in the two years following CT.

Methods: Inclusion criteria: children with CF; single center; presence of ≥1 routine bi-annual chest CT and ≥1 PFT obtained when clinically stable; 2 years follow-up data. RTE was defined as course of intravenous antibiotics. De-identified CT scans were randomized and scored by 2 observers with the Brody-II scoring system, assessing bronchiectasis, airway wall thickening, mucus plugging, opacities and trapped air. CT component scores (% max score) and PFT parameters (%-predicted) were divided into 4 categories (1) (2) (3) (4) with increasing degree of abnormality. Analysis included uni-and multivariate Poisson models for exacerbation count, intraclass correlation coefficients (ICC), and Bland-Altman plots.

Results: A total of 115 children contributed 170 CTs (55 M,60 F), 55 patients contributed 2 CTs. Median (range) age and FEV1 were 12 (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) years and 90 (23-132)%-predicted. Inter-and intraobserver variability was good (ICC range 0.61 -0.99). Number of exacerbations during 335 person years of follow-up was 138. Significant correlations were found between CT scores, PFT parameters and RTE-R. Bronchiectasis score and FEV1 were the strongest predictors for RTE-R in multivariate analysis. Exacerbation rate ratios for categories 2-4 relative to the reference category 1 were for bronchiectasis score 1.8 (0.6-5.7, p=0.30), 5.7 (2.1-15.5, p=0 .0006) and 8.5 (3.3-22.1, p<0 .0001) and for FEV1 1.5 (95% CI 0.6-4.0, p=0.42), 1.7 (0.6-4.9, p=0.33) and 4.7 (2.0-11.2, p=0.004) .

Conclusions: The severity of bronchiectasis on chest CT is a strong predictor of subsequent respiratory tract exacerbations in CF and is of additional value to FEV1. Rotterdam, Netherlands; 2. radiology, Erasmus MC, Rotterdam, Netherlands; 3. pediatric pulmonology, Cystic Fibrosis Centre, Verona, Italy; 4. epidemiology and biostatistics, Erasmus MC, Rotterdam, Netherlands; 5. Activaero GmbH, Gauting, Germany Introduction: A hallmark of respiratory disease in cystic fibrosis (CF) is obstruction of the small airways. During CF exacerbations there is more inflammation and sputum which results in a more central deposition pattern of inhaled medications in the lung.

Hypothesis: a more peripheral lung deposition of rhDNase during an exacerbation results in a better resolution of lung function abnormalities.

Aims: To compare the efficacy of rhDNase targeted to the peripheral airways with that of conventional more central airway targeted rhDNase in CF patients during a pulmonary exacerbation.

Methods: We performed a randomized controlled, double-blind, clinical trial in 2 CF centres and included 34 CF patients admitted to the hospital for a pulmonary exacerbation.

After a 5-day run-in period using a conventional nebulizer, patients were randomized on day 6 to rhDNase targeted to peripheral airways or to central airways, using the Akita ® nebulizer. Peripheral setting: MMAD 3.0 µm, slow inhalation with aerosol bolus at start of each breath. Central setting: MMAD 6.0 µm, normal inhalation with aerosol bolus in middle of each breath. Minimal study treatment duration was 7 days. Study treatment was continued until discharge from the hospital. At study days 1, 3, 5, 6, 12 and at day of discharge spirometry was performed. At day 2, 5, 6 and 12 nightly oxygen saturation profile was recorded. Primary endpoint is FEF 75 , secondary endpoints are FEV 1 , FVC, and nightly oxygen saturation parameters.

Statistical methods: We performed an ANCOVA-analysis for FEF 75 , FEV 1 and FVC, comparing the two treatment groups. Spirometry data were log-transformed and change in lung function over the 7-day treatment period was calculated for each patient. Difference was expressed as % change from baseline:(PFT day 12 -PFT day 5)/PFT day 5.

Results: Preliminary analysis showed that patients in the peripheral treatment group on average had a 10% bigger improvement in FEF 75 than patients in the central treatment group. However, this difference was not statistically significant: p = 0.49 and 95% CI of the change in FEF 75 of -15% to + 45%.

For FEV 1 the mean improvement in the peripheral group compared to the central group was 4%, p = 0.29 and 95% CI -4% to +13%.

Conclusions: Preliminary analysis of pilot data suggests that there might be a positive treatment effect of rhDNase targeted to the peripheral airways compared with rhDNase targeted to the central airways in CF patients admitted to the hospital for treatment of a pulmonary exacerbation. Since the 95% CI has a wide range, it is difficult to draw final conclusions from this study. For control of cystic fibrosis (CF) lung infections, treatment marker substances are needed. A good marker has to meet the following conditions: noninvasive and repeated collection, up-to-date reflection of the patient status, inexpensive and straightforward technology, extensive applicability. We investigated lactate in sputum of CF patients for its suitability as a biochemical marker for the outcome of CF lung exacerbation treatment.

Methods: In sputum supernatants of 25 adult CF patients without exacerbations from Halle, lactate concentrations were determined spectrophotometrically. In another 10 patients with exacerbation from Halle and 11 patients with exacerbation from Firenze, lactate was determined before and after IV antibiotic treatment. In all patients, lung function parameters FEV 1 and FVC were measured at the time of sputum expectoration using the Viasys Jaeger bodyplethysmograph (Halle) or the portable ZAN 100 morgan spirometer (Firenze).

Results: In every single CF sputum, lactate was detectable in a range from 0.2 to 14.1 mmol/L. In the stable CF patients, lactate values correlated negatively with FEV 1 (r=-0.446, p=0.043), whereas for FVC, correlation coefficients were lower. In the Halle exacerbated patients, lactate decreased from 4.8 ± 2.7 at exacerbation to 1.8 ± 1.6 mmol/L after treatment (p=0.003). The decrease in lactate correlated with FEV 1 (r=-0.467) and FVC (r=-0.725) increases. In the Firenze patients, the respective values were 2.8 ± 2.4 vs 1.1 ± 0.9 for lactate before and after treatment, and r=-0.143 for FEV 1 and r=-0.496 for FVC.

Conclusions: Thus, in two independent CF patient populations with two different types of lung function measurements we could show that amelioration of lung function values correlates with a decrease in lactate.

Lactate concentrations in CF sputum can be determined non-invasively and several times a day. It may be used for treatment control of stable as well as exacerbated CF patients. It is inexpensive and can be investigated using routine equipment. Antibiotic as well as antiinflammatory or even mucolytic therapies most probably are reflected by changes in lactate values. This is especially true for the rapid control of exacerbation treatment. Therefore, lactate is a promising marker substance for the short-term control of the progression of CF lung infection and inflammation. Exercise is increasingly prescribed for patients with cystic fibrosis (CF) as adjunct to chest clearance regimens, to increase the quality of life, and as a strategy that may prolong life. Previous studies have demonstrated desaturation during exercise in patients with CF when compared to healthy control subjects. The diffusing capacity of the lungs represents the transfer of gas from the alveoli to the pulmonary capillaries, is influenced by alveolarcapillary membrane conductance (DM) and pulmonary capillary blood volume (Vc), and is a factor that influences peripheral oxygen saturation (SaO 2 ). We sought to determine the influence of CF on DLCO, DM, and Vc along with SaO 2 during exercise. To determine this we recruited six mild/moderate CF patients (age=24±7years, ht=174±1cm. wt=67±6kg, BMI=22±2kg/m 2 , VO 2 peak=50±2%pred., FEV1=61±21%pred.) and 10 healthy control subjects (age=30±10years, ht=176±10cm. wt=76±13kg, BMI=25±5kg/m 2 , VO 2 peak=97±26%pred., FEV1=92±15%pred.) for study. Subjects performed cycle ergometry to exhaustion using three minute stage increments with continuous monitoring of SaO 2 and measures of DM and Vc taken at each stage using the diffusing capacity of the lungs for carbon monoxide and nitric oxide. There were no differences in SaO 2 at rest, however, the CF patients had a lower SaO 2 at peak exercise (SaO 2 rest=97±1 vs. 96±2%, SaO 2 peak=97±1 vs. 94±2, for healthy and CF, respectively, mean±SD, p<0.05 at peak). At rest, there were no significant differences in DM, Vc, or DM/Vc (alveolar-capillary conductance corrected for the amount of pulmonary capillary blood)(DM=36±7 vs. 32±5ml/min/mmHg, Vc=75±37 vs. 55±26ml, DM/Vc=0.59±0.2 vs. 0.63±0.2, for healthy and CF, respectively). At peak exercise, the healthy subjects demonstrated a significant increase in DM and sustained DM/Vc, whereas the CF subjects had no change in DM and a drop in DM/Vc (DM=53±14 vs. 34±8ml/min/mmHg, Vc=100±33 vs. 92±34ml, DM/Vc=0.56±0.15 vs. 0.39±0.11, for healthy and CF, respectively, p<0.05 healthy vs. CF for DM and DM/vc). At peak exercise, there was a mild relationship in DM/Vc and SaO 2 in CF patients that did not reach statistical significance, likely because of the relatively small sample size (CF r=0.51, healthy r=0.15, p=0.15 for CF and p=0.7 for healthy). These results suggest an attenuation in the increase in alveolar-capillary membrane conductance with exercise in patients with CF which may contribute, in part, to a drop in SaO 2 along with the ventilation/perfusion mismatch that is known in this patient population. Predictors of survival can be useful in planning treatment, especially when referral for transplantation is considered. The BODE index (Body mass index, degree of airflow Obstruction, Dyspnea, and Exercise capacity) is now accepted as a useful tool to predict survival in COPD. However, it has not previously been used to assess survival in patients with cystic fibrosis (CF). To study this further, we looked at a modified BODE index (mBODE) in a cohort of 65 adult CF patients (mean age 27 (range 18 to 46), 35 female) and compared its utility with the more standard measures of FEV1 and exercise tolerance in predicting survival.

As regards the BODE index, the values for BMI were taken as BMI <18 score=2 and BMI>18 score=1. FEV1 (% predicted) <30% score=3, 30% to <50% score=2, 50% to <80% score=1, and >80% score=0. Activity scale value 0.8 to <1.6 score=3, 1.6 to <2.4 score=2, 2.4 to <3.2 score=1, and 3.2 or greater score=0. Exercise capacity was assessed using a standard 6 minute walk test, where <500m score=3, 500m to 1000m score=2, and >1000m score=1. The cumulative mBODE score was then compared against the individual parameters as predictors of survival in all patients, using C statistic modelling (where the closer to unity, the better the model).

At the time of data analysis, the modified BODE index gave a mean score of 8 (range 3 to 12), mean FEV1 was 52% (range 17 to 110), mean BMI was 20.5 (range 15.6 to 28.8), median activity scale was 1.6 (range 0.8 to 4.0), and mean 6 minute walk distance was 536m (range 280 to 1440).

Nineteen patients (29%) had died (mean age at death 25 years): C statistics for mortality prediction were mBODE 0.78, BMI 0.66, FEV1 0.86, Dyspnea scale 0.54, and 6 minute walk distance 0.85.

Thus, based on this cohort of adult patients, although mBODE predicted survival better than nutritional state and symptoms, lung function and exercise capacity continue to be better predictors of survival in ill patients with CF rather than the multidimensional BODE index, perhaps reflecting the multi-system disease in these patients. Aims: The study aimed to select a clinically-significant definition of hypoxia in childhood CF.

Methods: Subjects were recruited from the Great Ormond Street Hospital CF clinic. Overnight oximetry (Minolta Pulsox 3i), spirometry (Jaeger Masterscreen), exercise testing with ventilatory gas analysis (MedGraphics), echocardiography, and measures of inflammation (full blood count)were performed. Data were entered into an SPSS v13.0 database. Receiver-operator (ROC) characteristics of existing definitions of sleep hypoxia were compared with new definitions based on percent of sleep time (5, 10, 15, 20, 25 , and 30%) spent with SaO 2 <93%, to ascertain the most sensitive and specific definition of sleep hypoxia in the detection of adverse outcomes; namely FEV1 and FVC <-2 z scores, peak VO 2 <32mls kg -1 min -1 , neutrophil count >8x10 9 ml -1 , systolic PA pressure >30mmHg, and the need for IV antibiotics. χ2 statistics (Fisher's exact test), as well as kappa scores (κ) for agreement between subject groups detected by hypoxia definitions and adverse clinical outcome criteria were used. Sleep hypoxia definitions were also compared with the EIAH definition to assess ROC characteristics for each outcome.

Results: Forty-one children (21 female) were studied. Median (range) age was 12.7 (8.0 to16.2) years, height -0.5 (-2.1 to +1.8) z scores, weight -0.7 (-2.2 to +1.7) z scores and FEV 1 -1.9 (-5 to +2) z scores. The hypoxia definition which optimally detected FEV 1 <-2 z scores was SaO 2 <93% for >10% sleep time, with an area under the curve (AUC) of 0.74 (χ2 12.1, p<0.001; κ 0.46, p<0.001). Similarly, SaO 2 <93% for >10% sleep proved best in detecting FVC<-2 z scores (AUC 0.76, χ2 11.3, p=0.002; κ 0.51, p=0.001), abnormal peak VO 2 (AUC 0.71, χ2 9.3, p=0.004; κ 0.46, p=0.002), elevated neutrophils (AUC 0.86, c2 15, p=0.001; κ 0.59, p<0.001), and the need for IV antibiotics (AUC 0.68, χ2 7.4 p=0.009; κ 0.36, p=0.006). When compared to the standard EIAH definition, hypoxia defined as SaO 2 <93% for >10% sleep time had a better sensitivity/specificity profile in detecting each adverse clinical outcome. Review of spirometry data suggests that an FEV 1 <70% predicted is 100% sensitive and 69% specific (PPV 47%) in detecting hypoxia defined as SaO 2 <93% for >10% sleep.

Discussion: The suggested optimal definition of hypoxia in children with CF is SaO 2 <93% for >10% sleep time. Applying this definition to our study group suggests 9/41 (22%) are hypoxic, yet only 1-2% children with CF receive oxygen therapy, suggesting potential under-recognition and under-treatment of hypoxia in CF. It is recommended that all children with EIAH undergo oximetry, and in those with FEV 1 <70% predicted, a hypoxic CF child would be detected on every 2-3 sleep studies. 4 ; Jaffe, A. 1, 5 1. Portex Unit, Institute of Child Health, London, United Kingdom; 2. Respiratory Medicine, Royal Hospital for Sick Children, Edinburgh, United Kingdom; 3. Cardiology, Great Ormond Street Hospital, London, United Kingdom; 4. Pediatric Pulmonology, HSC, Toronto, ON, Canada; 5. Respiratory Unit, Sydney Children's Hospital, Sydney, NSW, Australia Background: Hypoxia in CF may occur during sleep, exercise, chest exacerbations and air travel. Hypoxia is hypothesised to exert effects on growth, the pulmonary circulation, quality of life and also lung inflammation in CF, which may in turn have effects on lung function and exercise capacity.

Aims: To study the association of nocturnal hypoxia with clinical status in childhood CF.

Methods: Subjects were recruited from the Great Ormond Street Hospital CF clinic. Each child had overnight home oximetry (Minolta Pulsox 3i) before clinical testing. Hypoxia was defined as SaO 2 <93% for >10% sleep time in accordance with previous work. Spirometry (Jaeger), exercise testing with ventilatory gas analysis (MedGraphics), and echocardiography were performed, along with measures of inflammation: full blood count and serum IL-8 (BioSource). Quality of life was assessed using the UK CF questionnaire (CFQ-UK).

Results: Forty-one children were studied. Sixty-one percent were homozygous (32% heterozygous) for the F508del mutation. Median (IQR) values for clinical measures are displayed for hypoxic and normoxic groups ( Table 1) .

Discussion: Nocturnal hypoxia is associated with adverse growth, lung function, exercise capacity, X-ray scoring, cardiac function, inflammatory state and also quality of life in children with CF. Whilst hypoxia may be a causal mechanism for worsening clinical and inflammatory status, the paradigm exists that hypoxia may be a mere endpoint of worsening disease and that it is parenchymal lung damage as the final common pathway of airway inflammation which causes V/Q mismatch and resultant hypoxia. Future research should focus on elucidating effector mechanisms by which hypoxia may mediate CF lung disease, and also the extent to which restoration of normoxia (by oxygen therapy +/-non-invasive ventilation) impacts on the CF lung. Introduction: Systemic steroids and adjunctive antifungal therapy are the cornerstone in treating allergic bronchopulmonary aspergillosis (ABPA) in the context of cystic fibrosis (CF). Itraconazole and voricanazole are being used as antifungal drugs but absorption, drug interactions and/or side effects may limit their use and effectiveness.

Aim: Evaluate the use of inhaled amphotericin B (AmB) as antifungal agent in this context.

Method: A selection of 7 CF patients (see table) with recurrent or difficult to treat ABPA and failure to taper systemic corticosteroids were treated with AmphoB deoxycholate (AmB-D) (25mg 3x a week) or AmB lipid complex (ABLC) (50mg twice weekly). Data on IgE, lung function (LF) and medication use are presented. Successful therapy was defined as steroid withdrawal without ABPA relapse after 6 months.

Results: Therapy was successful in 6 of 7 patients treated with AmB-D or ABLC. In 4/6 lung function improved (1 patient too young for LF and 1 patient with B. cepacia infection progressing to respiratory failure). The patient with treatment failure has concomitant MAC lung infection.

Conclusion: Inhaled AmB may be an alternative to commonly used adjunct antifungal therapy in the treatment of ABPA. More data are needed on safety and efficacy. Background: Collection of Exhaled Breath (EBC) is non-invasive, fast and well tolerated method for collection of secretions from lower respiratory tract. Measurement of cytokines in EBC should reflect inflammatory changes in the airway surface liquid (ASL). However, some problems still remain, as concentrations of cytokines depend on method of EBC collection. The purpose of the present study is to compare cytokine levels in EBC to cytokine levels in sputum collected in the same patients on the same day.

Methods: Stable cystic fibrosis (CF) patients were asked to participate. EBC was collected breathing normally for 15 minutes using noseclip in an Ecoscreen® (Jäeger). Sputum was spontaneously expectorated before collection of EBC. Interferon-γ (IFN-γ), IL-1β, IL-6, IL-10 and TNF-α were measured using multiplex bead based immunoassay.

Results: In 32 CF patients (18 males), median age 27 years (range 14.2 -43.4 years) simultaneous samples of sputum and EBC were collected. All patients had chronic, gram-negative, pulmonary infection, and all patients accepted collection of EBC. Median FEV1, % of predicted, was 59.7 (range 30.8 -109.6).

Levels of cytokines in EBC: Levels of TNF-α, IL-1β, IL-6 and IL-10 were below detection limit (3.2 pg/mL) in all samples. IFN-γ was above detection limit in 14 of 32 samples, median 2.67 pg/mL (range 0 -9.34).

In sputum, levels were generally high: median concentration of TNF-α was 79.9 pg/mL (range 7.2 -960.9), of IL-1β 1685.4 pg/mL (range 35. 3 -16000) , of IL-6 53.2 pg/mL (range 0 -2213.5) and of IL-10 45 pg/mL (range 6.8 -291.6). Level of IFN-γ was below detection limit in 15 of 32 samples, median concentration 17.5 pg/mL (range 0 -190.3) . No correlation was found between levels of cytokine in EBC and lung function, age or levels of cytokines in sputum.

Conclusion: Collection of EBC is a safe, fast and non-invasive procedure, well accepted by the patients. Clinical usefulness of the method is questionable as results of EBC analyses vary between different centres and correlate poorly with sputum analyses. This method needs further standardization. Background: A growing number of adult CF patients choose to receive courses of intravenous (IV) antibiotics at home as treatment for pulmonary exacerbations. There is conflicting outcomes data comparing home-based IV antibiotic regimens with treatment courses that are solely hospital-based. At the Adult CF program of Brigham and Women's Hospital patients are admitted for 3-5 days and discharged home to complete their therapies if deemed eligible by their care team. Discharging such patients on home IV antibiotics requires a dedicated team to provide intensive home IV followup in order to ensure the highest level of care.

Methods: We retrospectively studied a cohort of 113 adult CF patients who received IV antibiotic therapy for pulmonary exacerbations over a 2 year period from 2006-2008. If a patient was discharged home to complete therapy, they received a guide that includes specific home-care instructions, contact information and were monitored with bi-weekly labs, phone calls and weekly clinic visits by a nurse practitioner or physician. Outcomes evaluated included the change in FEV1 between pre-and post-therapy, length of stay (LOS) and readmission rates within 30 days. We assessed differences in outcomes based on pre-admission disease severity.

Results: Over the 2-year period, 86 patients (mean age 30.4 ±7.9) completed 265 partially home-based IV therapies and 27 patients (mean age 33.4 ± 10.2) completed 61 hospital-based courses. The mean pre-admit FEV1 (53 ± 21% pred v. 53 ± 20% pred, p=0.97) and the percentage of patients with severe lung disease (FEV1<40%) in home-based (26.9%) and hospital-based (27.5%) were similar (p=0.929). In the home-based group, the LOS was 4.8 ± 3.1 days and was shorter compared to the hospital-based group (14.6 ± 7.6 days, p<0.001). The mean improvement in FEV1 following therapy between home and hospital-based IV was not significantly different (9.1 ± 10.7% pred v. 10.7 ± 15.5% pred, p=0.46). The percent 30-day readmission rates were identical for home-based and hospital-based therapy (9.8% v. 9.8%, p=0.9). The time to next admission for home-based was significantly longer compared to hospital-based (5.4 ± 4.8mo v. 3.2 ± 2.1mo, p<0.001) .

Conclusion: Home-based IV antibiotic therapy for CF pulmonary exacerbations is effective and leads to similar outcomes as hospital-based therapy in our cohort. There were no differences in treatment failure between the two groups (based on 30 day readmission rates). Our home IV program is a safe alternative in large part due to detailed patient education, close monitoring, partnerships with neighborhood home care networks and careful follow-up by an organized team of caregivers specifically dedicated to the home IV program. With newer nebulizing devices offering significant advantages in both time for pulmonary lung delivery and efficiency, confusion often arises with regard to the appropriate charge dose for various devices. The PARI LC Star ® breath enhanced nebulizer is one of the more efficient of its type and the newer vibrating membrane system, the PARI eFlow ® appears to be faster, more efficient and more flexible in that the particle size produced can be manipulated by changing the pore size in the membrane. The purpose of this study was to investigate relative equivalence of pulmonary deposition and time required for it. On 2 occasions, 6 normal adult males inhaled either 36 mg of normal saline in 4 mL using the LC Star ® driven by the PARI ProNeb Ultra compressor or 22.5 mg in 2.5 mL from an investigational model eFlow ® with a 30 membrane which produced a similar particle size distribution to the LC Star ® . The radiolabel 99m Tc-DTPA was added to both solutions. The choice of the charge "dose" was based on in vitro device performance which was used to estimate what would yield an equivalent in vivo lung deposition. Lung deposition was quantified using previously described nuclear medicine techniques (Coates et al. J Aerosol Med 2007; 20:320) which accounts for pulmonary clearance of the radiolabel during nebulization through both mucociliary action and absorption into the pulmonary circulation so the slower system will not be biased to favour the faster. Pulmonary deposition was 9.6 (8.8, 10.4) mg (mean with 95% C.I.) or 26.4 % of charge for the LC STAR ® and 8.6 (7.8, 9.4) or 38.0% for the investigational eFlow ® . The time to "dryness" of the LC STAR ® was 11.8 (10.1, 13.5) minutes and 4.8 (4.5, 5.0) for the investigational eFlow ® . There were no differences in pulmonary distribution for the two devices. The investigational eFlow ® is about 2.5 fold faster than the LC STAR ® . This study suggests that in vitro testing slightly overestimated eFlow ® efficiency. When changing from one nebulization system to another, it is essential to consider device efficiency when choosing the appropriate charge dose. Tamalet We have previously shown that one year of AZ treatment was associated with a decrease in respiratory exacerbations and number of antibiotic courses. The aim of this study was to confirm these results in a long term openlabelled trial in children with CF.

Methods: Fifty-seven children receiving AZ for at least 2 yrs (mean age: 10.7 ± 3.5 yrs) out of the 152 patients followed in our CF center were included. At months -12, 0, +12, +24 (month 0 being the start of AZ), the following data were extracted from the software e-muco: lung function parameters, Body Mass Index (BMI) z-score, sputum bacteriology, number of acute pulmonary exacerbations and of oral and intravenous antibiotic courses.

Results: The relative changes in lung function and BMI z-scores at months +12 and +24 were not different from month 0. The number of pulmonary exacerbations per patient was reduced from 3.9 ± 2.1 during the year prior to AZ to 2.6 ± 1.4 in the year after treatment (p = 0.004). This decrease was also observed for the number of oral antibiotic courses (3.5 ± 2.5 vs 1.4 ± 1.3, p= 0.003) but not for intravenous antibiotic courses. No change in Pseudomonas aeruginosa and Mycobacterium abcessus colonization was observed. Methicillin-sensitive Staphylococcus aureus colonization trended to decrease from 56% to 30% (p = 0.08), whereas methicillin-resistant Staphylococcus aureus colonization remained stable.

Conclusion: This open-label prospective study confirms the benefit of AZ in young patients with CF with regard to the decrease respiratory exacerbations and antibiotic courses.

Hutchinson, C.L.; Ferguson, K.; Orska, T.; Wyatt, H.; Price, J.F.; Ruiz, R.G. Paediatric Regional CF Unit, Kings College Hospital, London, United Kingdom

Regional heterogeneity of infection in cystic fibrosis (CF) lungs is well recognized. Targeted single lobe bronchoalveolar lavage (sBAL) may therefore miss infection which may be detected by more generalized multiple lobe bronchial washings (mBW) or endotracheal aspirates obtained after physiotherapy (pET) while awakening from general anaesthesia (GA). We compared the microbial yield from these three sampling methods.

Methods: All CF children who underwent fibreoptic bronchoscopy at our centre in 2007 and 2008 were included. Bronchoscopies were performed under GA via a laryngeal mask. sBALs were targeted at radiologically predetermined areas of abnormality or, in the absence of abnormalities, the right middle lobe. mBWs were then obtained by pooling sputa suctioned from both lungs without wedging the bronchoscope, instilling 2-3 ml saline into airways where necessary. Finally patients were intubated and the anaesthetic lightened to allow coughing. Physiotherapy consisting of positioning, saline instillation, manual hyperinflation, chest vibration and endotracheal suction was performed to obtain pET samples. S. aureus, P. aeruginosa, H. influenzae, S. maltophilia, M. catarrhalis, B. cepacia, S pneumoniae, MRSA, A. fumigatus and S. apiospermum were regarded to be respiratory pathogens. Any other cultured organisms and negative cultures were grouped together for this analysis.

Results: Over the two years 2007-8, 36 bronchoscopies were done for a variety of reasons in 34 CF children (19 female) with median age 9 years (range 2 months to 17 years). Comparing sBAL with mBW, 48 culture comparisons from 30 paired samples were made. For sBAL vs. pET, we made 55 culture comparisons from 32 paired samples. For mBW vs. pET, we made 44 culture comparisons from 32 paired samples. There was moderate agreement between sBAL and pET with κ=0.54 (p<0.0001). Agreement between both bronchoscopic samples sBAL and mBW was better with κ=0.60 (p<0.0001) and was best between the two generalized samples mBW and pET, κ=0.75 (p<0.0001). However, if only one or two sampling methods had been used, positive cultures for important pathogens would have been missed. P. aeruginosa would have been missed in 3 patients if sBAL were used alone, 4 with mBW alone and in 4 with pET alone. S. aureus would have been missed in 4 patients by sBAL alone, no patients by mBW alone and 1 patient by pET alone. P. aeruginosa would have been missed in 1 patient if sBAL were done with mBW and in 1 patient if sBAL were done with pET. None of the patients with S. aureus would have been missed if sBAL were combined with one of the other methods.

Conclusion: Our data suggests that generalized sampling should be combined with targeted sampling to improve the sensitivity of detecting pathogens in CF lower airways. We would recommend using all three sampling methods to obtain the maximum sensitivity. Introduction: Because of the potential pulmonary, nutritional and psychosocial benefits associated with early diagnosis of CF, the CDC has encouraged all states in the U.S. to implement newborn screening (NBS) and almost all are now doing so. In addition, the CDC recommends the continued collection of data regarding the outcomes of children diagnosed by NBS because uncertainty remains whether routine NBS programs will provide similar medical benefits proven in the research setting, due to differences in screening procedure and follow-up treatment among NBS programs. The Wisconsin randomized clinical trial of CF NBS (RCT), with enrollment from 1985-1994, led to the implementation of the Wisconsin Routine CF NBS (ROUTINE) program in 1994. Our recent analysis showed that early childhood growth in children diagnosed through routine statewide CF NBS program is comparable to that observed in the RCT setting.

Objective: To evaluate whether children diagnosed through the ROU-TINE program displayed similar lung disease outcomes observed in the RCT-screened cohort during the first six years of life, after 10 years of routine screening.

Methods: Seventy-five children diagnosed through the Wisconsin ROUTINE NBS program from 1994-2003 and followed in the Madison (n=35) and Milwaukee (n=40) CF centers were evaluated. Chest radiographs (CXR) obtained at diagnosis, 2, 4, 5 and 6 years of age (299 films) were scored by two independent pulmonologists using both the Brasfield System and the Wisconsin Chest Radiograph Scoring System, which is more sensitive in detecting early signs of lung disease. The average CXR scores were compared to those of 124 children enrolled in the Wisconsin RCT (Screened = 66, Control = 58, Total films = 475).

Results: The ROUTINE cohort showed significantly lower (better) Wisconsin CXR scores than the RCT-Screened (p = 0.033) and the RCT-Control (p <0.001) over the first 6 years of life, with greater differences observed during the first 4 years of life. No significant differences were observed in Brasfield CXR scores between the ROUTINE and the RCT-Screened (p=0.38) or the RCT-Control (p=0.23) groups.

Conclusion: Children diagnosed through the Wisconsin Routine NBS program during 1994-2003 show significantly less lung disease in the first 4 years of life, as revealed from chest radiographic scores, than those diagnosed in the RCT setting during 1985-1994, but these differences diminished between ages 4 to 6 years. Advances in pulmonary treatment during the past decade may contribute to these differences. Further analysis comparing pulmonary infections, in particular P. aeruginosa and S. aureus, between these two cohorts will be presented.

(Supported by NIH Grants R01DK072126, R01DK34108.) Introduction: Cystic fibrosis (CF) patients are frequently admitted to the hospital for prolonged IV antibiotic therapy to treat acute exacerbations. Peripherally inserted central venous catheter (PICC) and Implantable Venous Access Systems (IVAS) or ports are commonly used for drug delivery in these patients. Current literature provides limited and conflicting data regarding the likelihood of hypercoagulability in these patients.

To report a series of cases with intravascular device (IVD) or intravascular catheter (IVC) related non-infectious complications in a cohort of patients with adult cystic fibrosis.

Methods: A database was constructed reviewing the electronic medical records from 1995 to 2009 of all adult CF patients currently enrolled in our CF Affiliate Center Clinic. Data collected included date of insertion of device/catheter, date and type of complication, study used for detection of venous thromboembolism and hypercoagulability work-up.

Results: A total of 25 patient charts were reviewed. Eight out of 25 patients had intravascular device/catheter related non-infectious complications. Six of these 8 patients had upper extremity deep vein thrombosis (DVT), including subclavian thrombosis, detected by upper extremity vascular doppler ultrasound study or venography. One had extensive superior vena cava thrombosis and femoral port thrombosis. Two out of 9 patients had fracture of the intravascular component of the port. One of them had embolization of the fractured fragment of the port to the lobar branches of the pulmonary artery. Two out of 9 patients had port malfunction requiring removal and insertion of a new port. One of the patients with upper extremity DVT had Factor V Leiden mutation.

Conclusion: The prevalence of upper extremity DVT, port fracture, and port malfunction is high in adult CF patients.

Clinical point: A systematic approach looking for IVD or IVC related complications and thrombophilia is indicated in patients with cystic fibrosis. Later survival in these patients suggests the possibility of more prolonged indwelling catheter duration, for which the safety has not been established. There is a need for consensus guidelines for the prevention of thrombosis associated with prolonged indwelling catheters in CF patients. Objective: Airway hyperreactivity and bronchodilator responsiveness are common in patients with cystic fibrosis (CF); little is known about the prevalence of bronchodilator responsiveness in CF infants. Our objective was to determine prevalence of bronchodilator responsiveness in a population of CF infants who underwent clinically indicated infant pulmonary function testing (PFTs).

Methods: This IRB-approved retrospective study was conducted using the infant PFT database at the University of North Carolina Chapel Hill. Subjects were included in analysis if they had undergone pre-and postbronchodilator infant PFTs, carried a diagnosis of CF, and were not in the midst of a pulmonary exacerbation at the time of testing. PFTs were performed using the raised volume rapid thoracoabdominal compression technique to measure forced vital capacity (FVC), forced expiratory volume in 0.5 seconds (FEV 0.5 ), and forced expiratory flows at 75%, 85%, and 25-75% of FVC (FEF 75 , FEF 85 , and FEF 25-75 , respectively). Forced expiratory maneuvers were performed before and 10 minutes after 2-6 puffs of albuterol MDI (administered until subject's heart rate increased ≥10% or to a maximum 6 puffs). Bronchodilator responsiveness was defined in two ways: 1) increase in FEF 75 of >24% and/or increase in FEF 85 >39%, or 2) increase in FEV 0.5 of > 12.8% and/or increase in FEF 25-75 >24% (Goldstein et al. Am J Resp Crit Care Med 2001; 164:447) .

Results: Final analysis included PFTs on 22 CF infants aged 20-133 weeks (mean 83 wks); 50% were male. Ten infants (45%) were bronchodilator responsive by FEF 75 /FEF 85 criteria, while only 4 (18%) met FEV 0.5 /FEF 25-75 criteria. There was no significant difference in gender, subject age, or baseline (pre-bronchodilator) pulmonary function between bronchodilator responsive and non-responsive groups by either definition. Changes in FEV 0.5 , FEF 50 , FEF 75 , FEF 85 , and FEF 25-75 were significantly different between the bronchodilator responsive and non-responsive groups by both definitions; change in FVC was not significantly different between groups. Among the 10 infants with bronchodilator response by FEF 75 /FEF 85 criteria, FEV 0.5 , FEF 75 , FEF 85 , and FEF 25-75 increased by a mean of 8.9, 30.8, 39.3, and 22.3%, respectively, compared to increases of 4.0, 5.8, 7.9, and 4.3% in the non-responsive group (p<0.005). The 4 infants with bronchodilator response by FEF 0.5 /FEF 25-75 criteria had mean increases in FEV 0.5 , FEF 75 , FEF 85 , and FEF 25-75 of 10. 2, 35.3, 47 .0, and 30.6%, respectively, compared to increases of 5.4, 13.2, 16.6, and 8 .5% in the non-responsive group (p<0.008).

The prevalence of bronchodilator responsiveness in CF infants in our population was 45% when defined by change in FEF 75 and/or FEF 85 . This is substantially higher than the 21% responsiveness Goldstein reported in healthy controls (2001), but less than the 62% responsiveness previously reported in CF infants (Goldstein Pediatr Pulmonol 1999) . The 18% prevalence of bronchodilator responsiveness by FEV 0.5 and/or FEF 25-75 criteria is consistent with the 18% responsiveness seen by Saito in a group of infants with recurrent wheeze (Pediatr Pulmonol 2006; 41:709) . Higher prevalence of bronchodilator responsiveness in FEF 75 and FEF 85 compared to healthy controls may be indicative of early peripheral airways disease that is not reflected in changes in FEV 0.5 or FEF 25-75 . Harrison, A.N.; Castro, G.; Dashefsky, L.; Nussbaum, E. Pediatric Pulmonary, Allergy, and Cystic Fibrosis, Miller Children's Hospital, Long Beach, CA, USA Rationale: Infant pulmonary function testing (IPFT) has become a standard clinical tool for the evaluation of lung function in infants with cystic fibrosis (CF) at many large CF Care Centers. However, most centers utilize a conscious sedation protocol that uses chloral hydrate (CH) as the primary sedative agent. Due to several recent reports of unpredictability of both treatment effect and side effect profile, our pediatric sedation team elected to utilize continuous propofol infusion (CPI) instead for outpatient testing. Chloral hydrate was used for testing patients tested as inpatients. Our hypothesis was that CPI would provide comparable data to patients studied via a CH protocol.

Methods: A comparative analysis was performed of all young children with cystic fibrosis studied with IPFT at our institution from May 2008 to the present. A total of 9 patients were studied (3 with CH, 5 with CPI, and 1 with ketamine/versed secondary to soy allergy). Chloral hydrate was administered as an initial 50 mg/kg dose, followed by 25 mg/kg doses up to a total of 100 mg/kg, if necessary. Doses used for testing ranged from 50-100 mg/kg. CPI was performed with continuous intravenous infusion of propofol as needed. Doses used for testing ranged from 60-250 mcg/kg/minute of propofol, with most patients requiring doses at the lower end of the reported range. The average age at testing was 17.6 months. Patients were kept NPO for a minimum of 6 hours prior to testing, as per the policies and procedures in place at our institution. IPFT was performed using the Infant Pulmonary Laboratory by nSpire. Lung mechanics, including Rrs and Crs, were measured by the occlusion technique. Forced expiratory flows, including FVC, FEV0.5, FEF 25-75, FEF 75, and FEF 85, were obtained by raised volume rapid thoracic compression (RVRTC). Plethysmography was performed to obtain total lung capacity (TLC), residual volume (RV), and RV/TLC ratio. All patients were either observed closely in the pediatric recovery room following testing until ready for discharge (following CPI) or returned to the pediatric ward with continuous cardiac and respiratory monitoring until they had recovered fully (following CH).

Results: To our knowledge, this is the first report of CPI in CF infants for IPFT. Valid, interpretable data was obtained for all patients tested, regardless of sedative medication utilized. Overall testing was completed more expediently with CPI, due to less variability in patient response to sedative medication. Recovery time was also less in patients receiving CPI. The success rate of data acquisition was significantly higher during CPI, with 100% (6/6) of patients tested with CPI or other intravenous agents completing post-bronchodilator testing compared with only 33% (1/3) of patients tested with CH. There were no serious adverse events, although one patient in the CPI group did require supplemental oxygen administration during testing.

Conclusions: CPI is a valid alternative to CH for sedation for IPFT. We conclude that CPI for IPFT is safe and effective when children of appropriate ASA class are selected, supplemental oxygen is delivered if necessary, and sedation and monitoring are done by an experienced pediatric intensivist.

Background: As lung function may be "normal" in younger children with cystic fibrosis (CF), more sensitive outcome measures are needed to facilitate development of disease-specific therapeutics early in the course of disease. We anticipate that computer-assisted assessment of quantitative CT airway (A) and air trapping (AT) measurements will be sensitive outcome measures for clinical trials in children with MCFLD.

Objective: To further validate quantitative A and AT measurements, test-retest reproducibility for these outcome measures will be studied in children with MCFLD.

Methods: Test-retest reproducibility for quantitative A measurements (LD, lumen diameter and TAD, total airway diameter, mm) and AT measurements (% expiratory volume (EV) A1, % EV A2) were studied in 12 of 36 children enrolled in the Novartis CF Natural History Study during baseline testing. These 12 children initially underwent low dose inspiratory spirometer-controlled spiral CT (SCSCT) scans of the chest. This was followed by limited inspiratory upper and lower lung zone SCSCT scans on 2 separate maneuvers. A low dose SCSCT expiratory chest scan was then obtained followed by limited upper and lower zone SCSCT expiratory scans.

Results: For all 12 subjects test-retest reproducibility was determined for inspiratory imaging LD and TAD, and expiratory imaging AT % EV A1 and % EV A2. Mean age for the 12 subjects was 11.7 ± 1.9 years, with 7 females/5 males, mean height 146 ± 10.1 cm, mean weight 40.3 ± 8.2 kg, mean % pred. FEV1 103 ± 12.4, mean Brasfield CXR score 22 ± 0.7. In the 12 subjects there were no significant differences by paired t-Tests between test 1 (T1) and retest 2 (RT2) values for upper zone (T1: LD 3.29 ± 1.31 mm, TAD 5.41 ± 1.64 mm vs. RT2: LD 3.30 ± 1.30 mm, TAD 5.40 ± 1.61 mm) and lower zone (T1: LD 4.34 ± 1.15 mm, TAD 6.56 ± 1.30 mm vs. RT2: LD 4.35 ± 1.13 mm, TAD 6.55 ± 1.29 mm) A measurements. In 9 subjects test-retest comparisons have been completed for upper and lower lung zone quantitative air trapping and are presented in Table 1 . Differences within subjects (T1 vs. RT2) were significantly smaller than differences between subjects by ANOVA.

Conclusions: There is good test-retest reproducibility for quantitative airway and air trapping measurements providing support for utilization of computer-assisted assessment of quantitative CT measurements during clinical trials in children with MCFLD.

Background: As lung function may be "normal" in younger children with mild cystic fibrosis (CF) lung disease (MCFLD), more sensitive outcome measures are needed to facilitate development of disease-specific therapeutics early in the course of disease.

Objective: To evaluate the sensitivity and feasibility of computer-assisted quantitative airway (A) and air trapping (AT) measurements during an observational multicenter two year natural history study in children with MCFLD.

Methods: Children with MCFLD have been enrolled and are to complete low dose spirometer-controlled spiral CT (SCSCT) imaging at baseline (T1), 3 months (T2), 1 and 2 years. Computer-assisted analysis of quantitative A and AT measurements are being assessed to determine if changes in these parameters are better early indicators of disease progression than pulmonary function tests, Brasfield chest x-ray (BCXR) scores, or visual CT scoring.

Results: Twenty-three of 36 subjects have completed T1 and T2 testing. Mean age is 11.9 ± 2.5 years, with 13 females/10 males, mean height 146 ± 13.5 cm, mean weight 41.2 ± 10.8 kg, mean % predicted (%P) FEV1 104 ± 11.8, mean %P FEF25-75% 103 ± 25.3, and mean BCXR score 22 ± 1.5. There were no significant differences for %P FEV1 (T1: 104 ± 11.8 vs. T2: 101 ± 16.2 %P) and %P FEF25-75% (T1: 103 ± 25.3 vs. T2: 97 ± 29.3 %P) for T1 and T2 testing. Differences within subjects (T1 and T2) were significantly smaller than between subjects for lumen diameter (LD), total airway diameter (TAD), and wall thickness (WT) measurements [T1: LD 5.93 ± 0.76 mm, TAD 8.68 ± 0.93 mm, and WT 1.37 ± 0.14 mm vs. T2: LD 5.99 ± 0.86 mm, TAD 8.75 ± 0.99 mm, and WT 1.38 ± 0.15 mm] by ANOVA (Table 1) . For quantitative AT measurements there were no significant differences by subject group or testing session by ANOVA (Table 1 ). In 22 subjects evaluated for AT, no significant differences were noted for % expiratory volume (% EV) A2 and A3 at T1 vs. T2 [T1 AT: 10.8 ± 16.5 % EV A2 and 4.0 ± 8.2 % EV A3 vs. T2 AT: 13.8 ± 17.5 % EV A2 and 4.7 ± 7.0 % EV A3, P > 0.05].

Conclusions: In the first 3 months of subject participation in this study there were no significant changes in quantitative A and AT CT indices and in spirometry indices in children with initial MCFLD. The current study should determine whether computer-assisted analysis of SCSCT indices are feasible, sensitive outcome measures compared to traditional PFT measures, BCXR score, and chest CT scoring in children with MCFLD. We plan to compare these results at baseline and 3 months during the 2009 NACF meeting. 

Eber, E.; Mitterhuber, A.; Oberwaldner, B.; Zach, M.

The six-minute walk test (6MWT) has been widely used in patients with chronic lung diseases. However, there is limited information on the relationship between the 6MWT and lung function parameters in cystic fibrosis (CF) patients. We compared walk distance with oxygen saturation (SaO 2 ), perception of dyspnoea and fatigue, and lung function parameters in this patient group.

Eighty-six CF patients (39 m, 47 f; age 20.4 ± 8.8 years) with a wide range of disease severity performed 6MWTs following a standardised protocol. SaO 2 was registered continuously with pulse oximetry; perception of dyspnoea and fatigue were assessed by a visual analogue scale (VAS; mm) prior to and immediately after the 6MWT. Baseline lung function parameters determined were FVC, FEV1, MEF50 and MEF25.

For the whole group, the mean ± SD walk distance was 646 ± 114m (males: 701 ± 90m; females: 600 ± 113m), baseline and minimum SaO 2 were 95 ± 3% and 90 ± 5%, respectively (p < 0.001), and SaO 2 after the 6MWT was 93 ± 7%. Dyspnoea and fatigue increased significantly (2.3 ± 2.2mm vs. 4.2 ± 2.5mm, p < 0.001; 2.6 ± 2.3mm vs. 4.7 ± 2.7mm, p < 0.001). There were significant correlations between walk distance and lung functions (FVC: r = 0.65, p < 0.001; FEV1: r = 0.62, p < 0.001; MEF50: r = 0.52, p < 0.001; MEF25: r = 0.42, p < 0.001), but not between walk distance and delta dyspnoea, delta fatigue, minimal SaO 2 , and delta SaO 2 . FEV1 did not correlate with delta dyspnoea and delta fatigue, but with minimum SaO 2 (r = 0.65, p < 0.001) and delta SaO 2 (r = -0.53, p < 0.001). The above correlations between walk distance and lung functions, and between FEV1 and minimum SaO 2 and delta SaO 2 were also significant for both the male and female subgroups, with the correlations for walk distance and both FVC and FEV1 being stronger in the female than in the male group. For both genders and for the whole group, the correlations between walk distance and lung functions tended to be stronger in patients < 18 years as compared to patients > 18 years of age.

We found that walk distance is strongly correlated with baseline lung functions, but not with perception of dyspnoea and fatigue or oxygen saturation. In female patients, the correlations were stronger than in males, and in younger patients they were stronger than in older ones. It remains to be shown whether treatment-induced changes in walk distance are also correlated to the respective changes in lung function. Azithromycin is a macrolide having antimicrobial, anti-inflammatory and immunomodulatory properties without immune-suppression. Compared to oral and parenteral drug administration, inhalation may offer treatment benefits for several respiratory diseases (e.g. bronchiolitis obliterans, cystic fibrosis, chronic obstructive pulmonary disease, community acquired pneumonia, ventilator associated pneumonia, pulmonary ciliary dyskinesia). Hickey et al. (J Aerosol Med 2006; 19 :54) evaluated aerosolization efficiency of different concentrations of azithromycin solutions with jet nebulizers. The solutions were prepared from a reconstituted Zithromax ® freeze-dried powder. However, a reconstituted Zithromax ® solution is stable only for 7 days when stored under refrigeration and azithromycin is known for its very bitter taste. To improve patient-acceptance of nebulized azithromycin solutions, we developed novel, stable, taste-masked aqueous formulations with high azithromycin concentrations using excipients, which are generally regarded as safe. The inhalation formulation was composed to achieve an antioxidative effect in vivo, improve airway surface liquid hydration and mucociliary clearance.

An Arrhenius extrapolation based on 12-month storage stability data suggests the novel azithromycin formulations have a minimum shelf-life stability of 3 years when stored at 5∞C. Nebulisation experiments with customized eFlow ® configurations showed delivered doses of 60 to 75% of the nominal dose and fine particle fractions from 60 to 80%, independent from the azithromycin concentration (50-100 mg/mL). Inhalation tests confirmed good tolerability and acceptable taste. These results qualify the novel azithromycin solutions for pulmonary administration via eFlow ® technology platform devices configured for home and hospital use, and for (para-)nasal delivery via VibrENT™. Aims: The main objective of this randomized multicenter trial was to statistically analyze the efficacy of two early Pa infection-eradicating treatment protocols (inhalatory colistin/oral ciprofloxacin (arm A) versus inhalatory tobramycin/oral ciprofloxacin (arm B) for 28 days in a sizeable number of CF patients. The treatment was considered efficacious if patients were infection-free with at least three cultures negative for Pa over 6 months.

Results: A total of 136 patients (63 F and 73 M, mean age 8.8±7.3 yrs, median age 7, range 1-36) from 12 centers were enrolled. Sixty-two of 136 (45.6%) patients received inhalatory colistin/oral ciprofloxacin and 74/136 (54.4%) inhalatory tobramycin/oral ciprofloxacin. Seventeen patients withdrew from the study (7 in arm A and 10 in arm B). At present a total of 77/136 patients have completed the 6-month follow-up, 36/77 patients in arm A and 41/77 in arm B.

Pa eradication according to intention-to-treat analysis was achieved in 13/36 patients in arm A and in 23/41 patients in arm B. Interim analysis showed no statistically significant difference in the two arms of the study and boundaries (calculated according to O'Brian-Fleming's method) were not exceeded.

Microbiological studies showed that the Pa strains were susceptible to colistin, tobramycin and ciprofloxacin.

Currently, 63 baseline sera have been tested for the presence of anti-Pa lipopolysaccharide antibodies (ELISA kit for quantitative assay of IgG and IgA) above the cut-off at enrollment. Prior to bacterial isolation, antibodies higher than the cut-off values were found in 34% of patients.

Conclusions: At this point in the study we cannot reject the null hypothesis regarding absence of differences between the two types of treatment in relation to primary outcome. And up until now, cultured strains have been shown to be susceptible to the study drugs used in eradication treatment. At the moment antibody titers above the cut-off might be considered a marker to perform cultures more frequently to detect the presence of Pa in our patients' respiratory tracts.

This work was supported by the Italian Cystic Fibrosis Research Foundation (grant FFC#17/2007) with the contribution of "Delegazione FFC di Vicenza".

We thank the following CF Centers and Microbiology Laboratories: Ancona, Brescia, Cagliari, Cerignola, Cesena, Firenze, Genova, Gualdo Tadino, Messina, Milano, Palermo, Roma Bambin Gesù. Purpose: HRCT is increasingly employed in the follow-up of pulmonary disease in cystic fibrosis (CF) due to the sensitivity in both early and more advanced disease. The aim of the study is to describe the characteristics and distribution of mucus plugging in children and young adults with CF and to compare chest radiography with HRCT.

Method and materials: Between January 2007 and December 2008 HRCT was performed on 31 CF patients (13 boys, 18 girls -mean age 13.5 years, range 6-18 years) followed at CF centre of Milan (334 patients under 18 years) to better characterize morphological changes of the lung disease. HRCTs were scored with Brody score and the score for mucus plugging was extrapolated for each lobe in all patients. Two pediatric radiologists re-evaluated chest radiography performed around the time of the HRCT (within 10 days) focusing only on the detection and location of mucus plugging. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of plain radiography in detecting mucus plugging was calculated using the HRCT findings as the reference standard.

Results: Out of 31 patients, 26 showed evidence of mucus plugging on HRCT according to Brody's score: 39.7% of lobes involving the "central lung" (finger-in-glove sign) and 66.7% of lobes involving the "peripheral lung"(centrilobular nodules), with a predominance in the right upper lobe (83.9%). Mucus plugging in peripheral lung was not detectable on chest radiography. The sensitivity, specificity, PPV and NPV of chest radiography for mucus plugging in large airways was 55%, 89%, 86% and 62% respectively. The sensitivity was higher in right upper lobe (69%), but very poor in right lower lobe (42%).

Conclusion: HRCT appears to be the most sensitive modality for the detection of mucus plugging, especially when present in peripheral lung which, in our series, is more frequently involved than central lung. Chest radiography is very poor in detecting peripheral mucus plugging whereas its sensitivity for central mucus plugging is better when HRCT score is high. Our data confirm the key-role of HRCT for better characterization of lung disease especially in detecting early signs of disease progression which cannot be seen on chest radiograph.

Male, 8 years: HRCT appearance of mucus plugging in the right upper lobe.

Fleet, J.; Guha, K.; Piper, S.; Banya, W.; Hodson, M.

Context: Azithromycin has become widely used in the treatment of cystic fibrosis (CF) patients, with randomized controlled trials documenting improvements in lung function and rates of infective exacerbations. However previous trials have been limited to a period of six months in adults and 12 months in children.

Objectives: To determine if the documented benefits in lung function and exacerbation rates can be observed in a non-trial population and if this effect is sustained up to two years after initiation of azithromycin.

Design and Setting: A retrospective longitudinal study set at the Royal Brompton Hospital in London, completing 2008.

Methods: Following a search of the CF database, 88 patients were identified who met specified criteria for inclusion in the study. The percentage expected FEV1 and courses of IV antibiotics received were recorded at yearly intervals at two years prior to and two years after azithromycin was initiated.

Results: The percentage expected FEV1 deteriorated by 0.96% during the first year and deteriorated by 2.89% in the second year but improved by 1.76% in the first year after starting azithromycin before declining again in the following year by 2.18%. Using a paired t-test the change in lung function at one year prior to and one year after initiation of azithromycin was statistically significant, P=0.002. The total deterioration two years prior to azithromycin initiation (3.85%) compared to the two years after azithromycin initiation (0.42%) was significant, P=0.018.

There were 94 courses of IV antibiotics in the first year of the study, which increased to 103 in the year prior to starting azithromycin. After azithromycin was commenced, the number of courses in the subsequent year was 106 and increased to 126 in the following year.

Conclusions: Azithromycin initiation resulted in significantly improved lung function during the first year, reflecting previous data in the clinical trial population. However this trend was not sustained past the first year of treatment despite an overall two year benefit. Azithromycin resulted in a lesser increase in exacerbation rates in the first year following initiation of treatment. Again this was not sustained into the second year. This study therefore highlights the need for randomized controlled trials to investigate beyond the six month and one year cut-off. 

Introduction: Measurement of Lung Clearance Index (LCI) following multiple breath washout (MBW) of inert gas is an emerging technique for tracking progression of cystic fibrosis (CF) lung disease. An advantage is the ability to perform measurements in all age groups. Posture is known to affect Functional Residual Capacity (FRC) and infants conventionally perform MBW supine. To allow comparison between age groups we investigated the effect of supine posture in children with CF and healthy controls.

Method: LCI was calculated from MBW of 0.2% sulphur hexafluoride (SF 6 ), performed to a standardised protocol. Subjects completed 3 MBW manoeuvres to obtain an average, followed by standard spirometry (Easyone, ndd, Switzerland). After lying supine for 30 minutes, measurements were repeated with this posture maintained. Recruitment is ongoing.

Results: Six children with CF (3 female, median [range] age = 9[5-13] years) were recruited with ten healthy volunteers (3 female, median [range] age = 12[6-15] years). Mean(SD) sitting LCI was higher in those with CF (8.33(2.69) vs. 6.45(0.40), p=0.04), however no difference was seen in mean(SD) FEV 1 z-score (-1.10(1.39) vs. -0.66(0.71), p=0.42). Comparing measurements after lying between children with CF vs. healthy controls, there was a greater mean(SD) change in FEV 1 (-0.66(0.54) vs. 0.33(0.77), p=0.04) and LCI (1.10(1.16) vs. 0.16(0.24), p=0.02), despite a comparable change in FRC (-24.65(11.39) vs. -26.55(10.97), p=0.4) .

Conclusions: A similar reduction in FRC was seen in CF and healthy control groups, consistent with previously published studies, however the greater change in FEV 1 and LCI seen in children with CF has not previously been reported. This has implications for detection and tracking of lung disease from infancy using LCI. Trials of interventions that span age ranges where measurement position is altered will need to take into account the effects on lung function. Rationale: Cystic fibrosis (CF) pulmonary disease is characterized by infection and inflammation. FEV 1 is the most commonly used parameter to evaluate lung disease severity. Low levels of IgG have been shown to correlate with better lung function in cross sectional evaluation. It is unclear whether longitudinal changes in IgG values correlate with longitudinal changes in FEV 1 . The aim of this study was to evaluate the correlation between change in serum IgG and change in FEV 1 cross-sectionally and longitudinally.

Methods: In pediatric CF patients in follow-up at the University Hospital of Leuven, biometry and lung function are measured three-monthly from age 6 years on and serum IgG values are measured yearly. Measurements taken on the same day in the years 2006 and 2008 were used for retrospective analysis. FEV 1 was expressed as %predicted using the Wang equation; weight(W) and height(H) were expressed as z-score (Cole) and serum IgG values were converted to z-scores using the laboratory's normal values for age. Bivariate correlation and paired t-test were used for analysis.

Results: Paired data from 66 children (41 boys, 25 girls) with median age of 12.0years (SD 3. (3) H zscores. Mean(SEM) differences (2008 minus 2006) were 0.21(0.07) for W z-scores (p=0.004), -1.65%(1.94%) for FEV 1 (p=0.40) and -0.19(0.25) for IgG z-scores (p=0.47); all by paired t test. Differences (2008 Differences ( minus 2006 in weight z-scores and IgG z-scores did not correlate significantly with differences in FEV 1 Conclusion: In cross-sectional evaluation, IgG z-scores do correlate with lung function severity expressed as FEV 1 % predicted. But in longitudinal evaluation the variability of changes in both parameters are large and the correlation is lost in this relatively small data set with a relatively short follow-up period.

Neville, M.E.; Liu, S.; Artis, C.; Dadey, E.; Gupta, R. Transave Inc, Monmouth Junction, NJ, USA Background: Liposomal amikacin (Arikace™) is a liquid dispersion of amikacin encapsulated in liposomes prepared from dipalmitoyl phosphatidylcholine and cholesterol. Early preclinical studies showed that (1) the inhalation of liposomal amikacin (Arikace™) produces sustained high concentrations of amikacin in the lungs of rats, and (2) repeated daily dosing with liposomal amikacin (Arikace™) results in the accumulation of foamy alveolar macrophages. The current study was designed to determine if these foamy alveolar macrophages exhibit normal macrophage functions, such as cytokine expression in response to endotoxin, phagocytosis and fungitoxicity.

Methods: Healthy adult rats were dosed with liposomal amikacin (Arikace™) (90 mg/kg/day) or 1.5% saline once daily via inhalation for 3 cycles with a 30-day on,30-day off dosing regimen over a six month period. Selected cohorts in the liposomal amikacin (Arikace™)-treated groups were euthanized at 24 or 72 hr after each 30-day dosing period or 24 hr after each 30-day recovery period. Rats in the 1.5% saline group were euthanized 24 hr after each 30-day dosing period or after each 30-day recovery period. Lung histology, cell counts and cell differentials in BAL fluid, concentration of amikacin in lung homogenates and BAL fluid and functional assays of phagocytosis, yeast killing, and cytokine production were performed using the macrophages collected at the time of necropsy.

Results: The daily inhalation of liposomal amikacin (Arikace™) deposited high concentrations of amikacin (greater than 3 mg/g) in the lungs of rats after each 30-day dosing period. Microscopic examination of lung sections showed the presence of large foamy macrophages in lung tissue and in the BAL fluid at 24 and 72 hr post dosing for all treatment cycles. Evaluation of the foamy macrophages isolated from BAL fluid showed that the yeast killing, cytokine expression and phagocytic functions were normal. At the end of the first 30-day recovery period, there were fewer and smaller foamy macrophages in the histological sections of the lungs and in the BAL fluid and the concentration of amikacin in the lung decreased by ten-fold to approximately 300 µg/g. Additionally, evaluations conducted after each subsequent recovery period showed that the macrophages exhibited normal cellular (phagocytosis, cytokine expression, etc.) functions.

Conclusions: The daily inhalation of liposomal amikacin (Arikace™) leads to the accumulation of macrophages in the lungs of rats that decreases at the end of each 30 day recovery period. These macrophages retain normal cellular functions at the end of each dosing and recovery period over the three treatment cycles. Background: There is currently no standardized tool to measure pulmonary exacerbations (PEx). Advanced measurement models (Item Response Theory; IRT), have been applied to PEx measurement in COPD (EXACT-PRO). This study reports preliminary analyses using IRT modeling to establish a PEx measure for CF.

Methods: IRT Models were compared using data from AIR-CF2, a safety and efficacy trial of Aztreonam Lysine for Inhalation Solution (AZLI) in 211 patients. Following a 4-week run-in with TOBI ® , patients received 4 weeks of AZLI or placebo and were followed for 56 days post-treatment. Both the Seattle (Rosenfeld et al. J Pediatr 2001; 139:359) and Akron (Kraynack, 2009) scoring systems were used to identify key signs and symptoms of a PEx based on adverse events (AE) during the study. An exacerbation was defined as a course of IV antibiotic treatment and 1 of 4 symptoms from the Seattle scoring system. Patient-reported symptoms (CFQ-R) were collected 5 times during the trial.

Results: Three 1-parameter IRT models were compared to evaluate how much information each contributed to identification of a PEx. The 1 st model included 6 signs and symptoms identified as unique predictors of a PEx using logistic regression (decline in FEV 1 >10%, increased cough, changes in sputum, increased dyspnea, hemoptysis, new chest sounds). The 2 nd model included 6 additional signs and symptoms (fatigue, wheeze) derived from AE records. The final model included decreases on 4 CFQ-R scales (Respiratory, Physical, Health, Vitality). Graphs of each model's information function are depicted in Figure 1 . Higher numbers indicate more information and greater precision. Model 1 provides the least amount of information about PEx events and Model 3 provides the most information, particularly for patients with mild to moderate symptoms. Results suggest 2 ways to increase the precision of a PEx tool: 1) include more signs and symptoms and 2) include PRO data.

Conclusions: IRT modeling suggests that patient-reported symptoms contribute important information to PEx scores, particularly when symp-toms are mild to moderate. Next steps include standardizing clinician assessment of chest signs and prospective evaluation of a tool combining PFTs, clinician-reported signs and PRO symptoms.

Funding by Gilead Sciences and the CF Foundation. Background: Levels of sputum cytokines may reflect the degree of inflammation in the cystic fibrosis (CF) airway, and thus be useful to monitor changes in response to interventions. However, many of the methods used for cytokine detection have not been formally validated for use with sputum. Dithiothreitol (DTT) has commonly been used to separate cellular components for total and differential cell counts. However, DTT acts by cleaving disulphide bonds, present in soluble mediators and in reagent antibodies; analysis of the soluble phase with immunoassays may therefore be significantly impaired in the presence of DTT. It has been previously recommended for non-CF sputum that DTT be removed by dialysis under conditions that encourage the reformation of disulphide bonds, as this improves cytokine detection. Here we looked at the effects both of DTT and its removal with dialysis on detection of cytokines with a Luminex system in CF sputum.

Methods: Spontaneously expectorated sputum was collected from CF patients during infective exacerbations. Sputum was processed according to a standard DTT protocol and soluble phase from different subjects pooled to give a total volume of 37 mL; aliquots were then treated with protease inhibitors and spiked with multi-cytokine standard (300-900pg/mL dependent on cytokine). Volumes (1mL) were transferred into sealed dialysis membrane sacks (Visking). Nine different dialysis buffers were tested containing different concentrations of arginine, guanidine and reduced:oxidised glutathione to which the samples were exposed for 20 hours; this was followed by a second 20 hours of dialysis in phosphate buffered saline. Samples were frozen prior to analysis. IL-1β, IL-6, IL-10, IL-12 (p40)/(p70), TNF-α, INFγ and RANTES were measured by the Luminex multiplex system.

Results: We confirmed that DTT does reduce the recovery of the cytokines IL6, IL10, IL12(p40)/(p70) and RANTES from DTT-containing buffer, but this was not due to an effect of DTT on the immunoassay capture antibody. In sputum samples, recovery of the spiked IL1β was improved by removal of DTT (range of recovery: 21-44% no dialysis, 33-97% with dialysis). In contrast, recovery of TNFα was reduced by dialysis, particularly when guanidine was included in the dialysis buffer. Recoveries of IL-6, IL-10 and RANTES remained low (<25%) following dialysis. Endogenous IL-1β concentrations were also increased by dialysis (63 pg/mL no dialysis, 495 pg/mL with dialysis); detection of no other cytokine was altered. Mod-ifications to the protocol (altering the type of protease inhibitor used and the length of time of dialysis) were tried but without significant improvement.

Conclusions: DTT impairs the detection of some cytokines in buffer and in soluble phase sputum. However, with the exception of IL-1β, the removal of DTT by dialysis did not lead to an increase in cytokine levels. DTT significantly decreased the recovery of IL-10, IL-6 and RANTES but this could not be improved by dialysis. Thus, we are unable to confirm that altering the redox environment encourages reformation of disulphide bonds. We would not recommend that such dialysis protocols be used for processing CF sputum samples.

Supported by the CF Trust. Recombinant human deoxyribonuclease (rhDNase) has been shown to improve pulmonary function and high resolution computerized tomographic (HRCT) imaging scores in children with cystic fibrosis (CF). The purpose of this study was to quantify the effects of rhDNase in infants with CF using infant pulmonary function tests (IPFTs) and controlled ventilation HRCT chest imaging. Twenty-four infants who were clinically well at the time of entry into the study were randomized to receive rhDNase (2.5 mg) or placebo by nebulization once daily for 6 months. Investigators and parents were blinded regarding treatment group. Raised volume rapid thoracic compression spirometry, plethysmographic fractional lung volumes and controlled ventilation HRCT images of the chest at 25 and 0 cm H 2 O were obtained at baseline and after 6 months of treatment. Lung function and HRCT measurements were obtained on 24 infants at baseline (mean age 42 ± 32 weeks) and on 19 and 21 infants respectively at 6 months. IPFT results were expressed as percents of predicted and as z-scores based on published normal values. HRCT images were quantified using the Brody scoring system modified for very mild disease. Airway wall thicknesses (AWT), airway lumen diameters (ALD) and accompanying vessel diameters (VD) were measured from axial images with electronic calipers using a GE workstation. Mean AWT/VD and ALD/VD ratios were calculated at baseline and at 6 months for each patient. Between treatment group differences in baseline to 6 month changes were assessed using unpaired t-tests. At baseline mean lung function was near normal (FEV0.5 98.1 ± 13.4, FEF25-75 87.4 ± 24.7 and RV/TLC 86.4 ± 23.4% of predicted). No significant differences in changes (∆) from baseline to 6 months between the two treatment groups were detected for any of the IPFT measures expressed as z-scores (z). Similarly no significant between treatment group differences were found for baseline to 6 month changes in total HRCT score, air trapping (AT) subscore, AWT/VD or ALD/VD ratios (see table below). In this relatively healthy population of infants with CF, we were unable to detect any differences between patients treated with rhDNase and placebo over a 6 month interval using IPFTs and HRCT imaging of the chest.

Study supported by funding from Genentech, Inc. 

Introduction: Bronchopulmonary changes of cystic fibrosis (CF), e.g., bronchiectasis and mucus plugging, have been well elucidated at CT in children; the most common distribution is upper lobe predominance and rightsided > left-sided disease. However, there is little information on CT findings for adult-diagnosis CF patients. We hypothesized that adult-diagnosis patients may differ from pediatric-diagnosis CF patients in extent and distribution of disease on CT imaging.

Methods: For 12 CF patients at our center diagnosed after age 18, the earliest available CT scan was obtained (5 mm sections; expiratory images not employed). Two thoracic radiologists independently evaluated the scans, using the Brody scoring system (J Pediatr 2004; 145:32-38) , and when differing, agreed by consensus on a final score. A third thoracic radiologist adjudicated unresolved discrepant scores. Location and severity of five radiologic findings (bronchiectasis, mucus plugging, peribronchial thickening, parenchymal findings and hyperinflation) were recorded. The lingula was scored as a separate lobe. Paired t-tests, Kruskal-Wallis, and McNemar's tests were used for comparisons among lobes. Median severity scores are reported.

Results: Median age at diagnosis was 33.5 (range 21-63) years. Median FEV1 was 76% predicted (range 42%-99%). Bronchiectasis and peribronchial thickening were the most common findings, in 71% and 69% of all lobes, respectively. Bronchiectasis was similarly prevalent on the right (27/36 lobes) and left sides (24/36 lobes), p=0.375. Bronchiectasis severity was also comparable between the right (score 24.5, range 0-40.3) and left sides (score 24.25, range 0-37.5), p=0.25. There was no significant difference between the upper lobes (score 21.25, range 0-31.75) and lower lobes (score 15, range 0-24.25), p=0.23. Among the individual lobes, bronchiectasis appeared to be more severe in the right upper lobe (score 4.5, range 0-10.5) compared to the left upper lobe (score 2.37, range 0-8), but this result was not statistically significant. Two patients had bronchiectasis and peribronchial thickening limited to a single lobe only.

Conclusions: The findings at CT scan of adult-diagnosis CF resemble typical CF disease, although distribution may possibly differ from the pediatric population. Bronchiectasis was the most common finding, but there was no predilection of disease for any specific lobe or side. If there is upperlobe and right-sided predominance, as has been previously reported in pediatric CF populations, it was not evident in this study, raising the possibility that adults manifest CF lung disease differently. Further case analyses are needed to increase the power to definitively detect differences between adult and pediatric-diagnosis patients. Late-presenting CF can manifest with atypical imaging, such as bronchiectasis and peribronchial thickening confined to a single lobe only. Awareness of the range of radiologic presentations in adults may allow for improved recognition of CF in symptomatic older patients presenting with bronchiectasis.

Navratil, T. 1 ; Schaberg, A. 1 ; Mathews, D. 1 ; Deans, C. 1 ; Durham, T. 1 ; Accurso, F.J. 2 1. Inspire Pharmaceuticals, Inc., Durham, NC, USA; 2. University of Colorado Denver, Aurora, CO, USA Introduction: Current treatment of cystic fibrosis (CF) involves a large number of inhaled, intranasal, oral and intravenous medications to treat pulmonary manifestations of the disease. In this abstract we describe concomitant medication use observed in mild CF patients randomly assigned to placebo in a large phase 3 clinical trial.

Aims: In order to evaluate the overall medication burden for CF patients, we have retrospectively examined concomitant medication use data for CF patients in Study 08-108, a clinical trial of denufosol tetrasodium inhalation solution. Demographic, baseline characteristics and concomitant medication use for 175 CF placebo-treated patients are reported.

Methods: A Phase 3, multicenter, randomized, double-blind, placebocontrolled, parallel-group trial was conducted at 61 CF centers in the Unit-ed States and 1 CF center in Canada. Eligible patients had screening FEV 1 % predicted of ≥75%, had no respiratory infections for at least 4 weeks prior to screening, and were allowed to use all current pharmacologic therapies available to treat CF except for hypertonic saline. Patients were randomized to receive either denufosol (60 mg) or placebo (normal saline) three times daily for 24 weeks by inhalation.

Results: From a total of 352 patients, 175 patients received placebo during the 24-week double-blind portion of the study. The mean age in the placebo safety population was 14.8 years with 45% of patients aged 5-11, 37% of patients aged 12-18 and 19% of patients aged 19 years and older. The mean FEV 1 % predicted was 92%, the mean FEF 25%-75% % predicted was 84%, and the proportion of patients with P. aeruginosa + cultures was 48%. At baseline, placebo-treated patients were using the following medications: selective short-acting or long-acting beta-2-adrenoreceptor agonists (84%), dornase alfa (77%), macrolides (41%) and chronic inhaled antibiotics (38%). During the study, 91% of the placebo-treated patients used selective short-acting or long-acting beta-2-adrenoreceptor agonists, 36% used fluoroquinolones, 27% used leukotriene modifiers, 26% used antihistamines for systemic use, 20% used high dose ibuprofen, and 18% used cephalosporins. In addition to these medications, placebo-treated patients received multiple additional pharmacotherapies including IV antibiotics, alternative mucolytics, inhaled and oral corticosteroids, and mast cell stabilizers. The percentages of placebo-treated patients on multiple concomitant medications (including inhaled) to treat pulmonary manifestations of CF will be presented. Further, the proportion of subjects receiving individual medications of interest, stratified by patients' total number of concomitant medications and total number of concomitant inhaled therapies, as well as the concomitant medication use in patient subgroups of interest, will be described.

Conclusion: CF patients face a large treatment burden. New prophylactic treatments are needed that address the downstream effects of CF disease, thus potentially reducing the need for extensive polypharmacy.

Acknowledgements: This research was funded by Inspire. The new PARI eFlow ® rapid allows a highly efficient aerosol delivery at reduced inhalation time. However, only rare lung function data during long-term use of this device are available until now.

Methods: Seventy clinically stable adult patients with cystic fibrosis (CF) participated in this observational study. Lung function tests were performed prospectively 12 weeks after and again 9 to 12 months after switching the inhalation device from a conventional jet nebulizer to the PARI eFlow ® rapid. Lung function data were collected retrospectively from the visits 1 year as well as 12 weeks prior to the switch-over. Lung function data for all time points were only available for 59 patients. Treatment time and patient's satification were recorded for both conventional and new nebuliser in all 70 patients.

Results: After 1 year of inhalation with eFlow ® rapid, the mean change in FEV1% was -1.4% (n=59 patients). The decrease in FEV1 was smaller than the change in FEV1 after 1 year of inhalation with the conventional jet nebuliser (control period, -3.1%), although this difference was not statistically significant. The same effect was seen in MEF25[%] (-2.6% with conventional nebuliser compared to -1.6% after eFlow ® rapid). Concerning the FVC, there was a greater improvement after 1 year of inhalation with the eFlow ® rapid than with the jet nebuliser (+ 2.9% vs. +1.1%). For PEF%, there was an increase during the control period, whereas after inhalation with eFlow ® rapid there was a decrease (+1.1% vs. -2.9%). All changes were not significantly different. The eFlow ® rapid reduced total daily inhalation time by two-thirds (conventional nebuliser: 31.1 min/day; eFlow ® rapid: 10.2 min/day, n=70 patients) Conclusion: Inhalation with the new nebuliser eFlow ® rapid does not alter FEV1, FVC or PEF significantly after 1 year of inhalation. The treatment time could be significantly reduced by inhalation with the eFlow ® rapid. Background: Pulmonary exacerbations compromise patient well-being and augment disease burden for patients with cystic fibrosis (CF). Lack of consensus regarding the definition of exacerbation complicates the evaluation of exacerbation frequency, severity, and duration in CF therapeutic trials. Further efforts are needed to standardize the assessment of exacerbations, particularly considering the patient perspective. Exacerbations in COPD and CF have common clinical features, including dyspnea, coughing, sputum production, chest discomfort, and fatigue. The EXACT is a patientreported outcome (PRO) measure designed to standardize the assessment of pulmonary exacerbations in clinical trials involving patients with chronic obstructive pulmonary disease (COPD). The EXACT total score is derived from 14 questions that capture information on Breathlessness (5 items), Cough and Sputum (2 items), Chest Symptoms (3 items) and systemic attributes of exacerbation (4 items). The EXACT improves accuracy and eliminates recall bias by asking patients to provide an evaluation of their health state, including pulmonary symptoms, every day before bedtime. Data entry and transmission are facilitated through use of a portable, handheld electronic device and require ~3 minutes to complete. This study examined the suitability of the EXACT for patients with CF by examining the extent to which items comprising the EXACT are understandable and meaningful to these patients.

Methods: In this single-site, observational study, adult and pediatric patients with CF completed the EXACT and participated in a cognitive debriefing interview about the tool.

Results: A total of 25 CF patients were interviewed, including 10 adults (ages 19 to 39 years), 5 adolescents (ages 12, 13, 14 [n=2], 17 years), and 10 children (ages 6 [n=2], 8, 9 [n=2] , 10 [n=2], 11 [n=3] years), with children participating in parent-child dyad interviews. Median [range] %-predicted FEV 1 was 76. 5 [26.0-138.0] ; median [range] number of pulmonary exacerbations (defined by medical records) in the prior 12 months was 3 [1] [2] [3] [4] [5] . Item content assessments provided by CF patients closely matched those of COPD patients. Participants in this study reported that all questions were relevant to CF pulmonary exacerbations; no additional concepts were reported. Patients ≥12 years old easily completed the EXACT; patients 9 to 11 years old needed parental assistance; and patients ≤8 years old needed a parent proxy to complete the EXACT for them.

Conclusions: The EXACT may offer a standardized method for documenting the frequency, severity, and duration of CF pulmonary exacerbations in the clinical trial setting based on patient-reported data. This study suggests the items comprising the EXACT are relevant to patients with CF across a range of ages, including both adults and children and provides guidelines for using the EXACT in young patients. In conjunction with physician assessment of exacerbations, the utility of the EXACT to document CF pulmonary exacerbations will be assessed in a Phase 3 study of ataluren (PTC124™) as a treatment for nonsense mutation CF. Introduction: Detecting a pulmonary exacerbation (PE) is important in outpatient cystic fibrosis (CF) care. Since the definition of a PE is subjective, identification of an exacerbation varies among practitioners, and this inconsistency has been identified at our CF Center. Scoring systems have been developed in other institutions to standardize identification of PEs in a more objective manner in order to improve care. Our Center has used a categorical assessment of pulmonary status, but we implemented a pulmonary exacerbation score (PES) in 2007 to improve pulmonary care and outcomes in our CF population. The objective of this study was to compare PES to the categorical system and to determine if implementation of the PES improved detection of an exacerbation and therefore increased treatment and followup.

Methods: This was a retrospective chart review. We selected patients followed at our CF Center based on their age, ability to perform reliable PFTs and the number of visits they had within the study period. We reviewed charts to obtain data from visits six months before and after the PES was implemented. Categorical assessment and interventions were recorded before and after April of 2007, the date that PES was instituted. Rate of interventions was also assessed based on BMI%, FEV 1 % and Pseudomonas status after the score was implemented. Data was analyzed with Chi-square and two sample t-test.

Results: We included 46 patients (21 males) in the study (Mean age=12.3). Eight patients did not meet criteria for number of visits or ability to perform reliable PFTs. After PES was implemented, providers' threshold to intervene was at lower scores and higher FEV 1 % than their threshold to categorize patients as "worse." This lower threshold did not result in a statistically significant increased rate of any of the interventions measured in the study; however, there was a trend toward increased use of intravenous antibiotics. Results also showed that providers did not perceive Pseudomonas positive patients as being subjectively sicker, although their PES scores and the rate of interventions was significantly higher than for Pseudomonas negative patients.

Conclusions: The implementation of an objective tool to detect exacerbations lowered providers' threshold to intervene, however, there was no significant change in the rate of any of the interventions after PES was implemented. This could result from an insufficient number of patients or an inadequate time frame, both of which would result in underpowering of the study. Also, the relative weighting of the sub-score elements of PES may have biased the results. PES may be a better tool than a categorical system to identify pulmonary exacerbations in patients with CF. Background: Cystic fibrosis (CF) is characterized by progressive destruction of the lungs with highly elevated levels of pro-inflammatory mediators -i.e. the chemokine IL-8 -leading to massive infiltration of neutrophils into the airways. In the CF airway microenvironment, airway neutrophils undergo apoptosis and necrosis and release significant amounts of pro-inflammatory mediators and neutrophil elastase (NE) resulting in further tissue damage. CF is caused by mutations in the CFTR protein which is the major transporter of the antioxidant glutathione (GSH) in airway epithelial cells. CFTR mutations are believed to lead to depleted levels of GSH in the epithelial lining fluids (ELF). It was hypothesized that due to extracellular GSH deficiency effector cells like neutrophils loose intracellular GSH into the ELF. Low intracellular GSH levels are associated with oxidative stress, inflammation and apoptosis of cells.

We aimed to systematically compare the characteristics of peripheral blood and airway neutrophils with particular focus on intracellular GSH levels and cell surface receptors.

Methods: Induced sputum and blood samples of CF patients were compared to samples of healthy subjects to investigate changes neutrophils may undergo with transmigration from blood into the airways. Sputum samples were processed by a mechanical method omitting dithiothreitol. Extracellular GSH was measured by RP-HPLC, intracellular GSH and neutrophil surface markers were assessed by FACS. Statistical analyses were performed with the non-parametric Kruskal-Wallis test.

Results: As known for sputum, extracellular GSH levels were elevated in CF patients compared to healthy subjects (p<0.001), but intracellular levels in non-apoptotic neutrophils were not different. Percentage of apoptotic neutrophils was significantly higher in CF population (p<0.001). In sputum, total cell count, percentage of neutrophils and NE levels were significantly higher in CF patients compared to healthy controls (p<0.01; p<0.05; p<0.001, respectively). CD35 was significantly higher in blood of healthy population compared to CF patients (p<0.05). CD63 showed elevated expression on airway neutrophils compared to blood neutrophils in CF patients (p<0.001), demonstrating activation of these cells, which was not found in healthy subjects. CXCR1 on neutrophils was depleted in sputum samples compared to blood samples in both populations (p<0.001) while CD11b was elevated in sputum samples in CF and healthy samples (p<0.05; p<001, respectively).

The major conclusions of this study in view of previous related investigations (1) are:

(a) Intracellular GSH levels in neutrophils were similar in sputum samples of CF patients and healthy subjects.

(b) Similar to previous reports, a translocation of CD63+ primary granules to the cell surface membrane was found.

(c) Cleavage of CXCR1 in airway neutrophils was found consistently in CF patients, but was also found in healthy subjects.

This study was supported by CF Foundation Therapeutics, Inc., USA and Mukoviszidose e.V., Germany. Cystic fibrosis (CF) is the most common life-shortening inherited disease, particularly among Caucasian populations, but now is a pan ethnic disease.

Current guidelines recommend that the health care of CF should be delivered by a specialist team in a CF Clinic or centre. In emerging and developing economies there are some problems to manage a complex disease such as CF: one of them is the lack of specialized clinics to appropriately treat these patients. In 2001, our CF clinic began to work and received patients previously diagnosed who had not been followed in specialized CF clinics. Since 2001 we diagnosed and followed new patients. The management of both groups of patients gave us the opportunity to study and compare the effects of specialized care.

Objectives: To assess the effect of CF treatment on clinical outcome, delivered at a specialized clinic vs. outcomes at historical control group of age-matched CF patients not treated on a specialized clinic.

Subjects: Patients younger than 18 years old were divided in two groups: A) patients that were born between 2000 and 2009 and followed at the CF Clinic since their first year of diagnosis; and B) patients who had received care in other centres not specialized in CF until they arrived at our clinic, born between 1990 and 1999.

The characteristics of two groups of CF patient are Group A (n=26), female 13 (50%); CF genotype 508/508 : 10(38%), 508/other: 8 (31%), non-508/non-508: 6 (23%), Neg. 2 (8%); Liver disease: Cirrhosis: 2(8%), Elevated enzymes: 2(8%), Gallstones: 1 (4%); Mucoid Pseud. aerug.: 4 (15%); B. cepacia complex: 0; Other resistant microorg.:0; Death:0. Group B (n=21), female 11 (52 %), male 10 (48%); CF genotype 508/508: 8 (52%),508/other: 12 (57%), non-508/non-508: 1 (5%); Liver disease: Cirrhosis: 8 (38%), Mucoid Pseud. Aerug.: 16 (76%); B. cepacia complex: 8 (38 %); Other resistant microorg.: 8 (38%); Death: 15 (71%), Age of Death 9 + 1.3, related to lung disease 12, other causes:3.

Main Outcome measures: Body Mass index, Lung Function (FVC and FEV1 percentage of predicted) at age 4, 6, and 8 years old, X-ray score (Brasfield), mucoid P. aeruginosa, Burkholderia cepacia complex and other resistant bacteria that accelerated lung function decline at the first culture in our clinic, Oxygen saturation at rest as measured by pulse oxymeter.

Results: are shown in Table 1 . Conclusion: In these two groups of CF children: 1) Clinical outcomes were better when treated at a specialized center.

2) Remarkable differences were found on respiratory resistant microorganism presence and deaths.

In emergent and developing economies is imperative to establish CF Services for better outcome. (HTS) has been shown to be an inexpensive, safe, and efficacious treatment for reducing the frequency of pulmonary exacerbations and improving quality of life in cystic fibrosis (CF) patients with moderately reduced lung function (FEV1 >40% predicted). However, HTS may precipitate bronchospasm and promote swelling of intraluminal mucus plugs that can be difficult for patients with severe lung disease to mobilize. The purpose of this pilot study was to evaluate the safety and tolerability of HTS in adult CF patients with severe lung disease.

Methods: During a 5 month period, adult CF patients with baseline FEV1 <40% predicted were identified at 2 CF Centers. After meeting study criteria, participants signed consent and enrolled for testing. Baseline spirometry was obtained after pre-treatment with albuterol via metered dose inhaler. If the FEV1 was <40% predicted, 5 mL 7% HTS was administered by nebulizer, and spirometry was repeated 15 minutes later. During and after HTS treatment, study subjects were closely monitored, and each was given a questionnaire to identify subjective findings related to treatment. Spirometric values pre-and post-HTS were compared using the two tailed t-test.

Results: Twelve adult CF patients were enrolled and 11 completed the testing visit. One patient was excluded from treatment due to hypoxemia prior to testing. Pre-HTS mean FVC was 2.37 L (range 3.65-1.24L), mean FEV1 was 1.05 L (range 0.71-1.55L) and mean FEV1 % predicted was 30%. Spirometry post-HTS revealed little change overall: mean change in absolute FEV1 was -1.26% (p=0.71, SD 0.11) and the mean difference in FEV1 % predicted was -0.73 (p=0.44; SD 2.9). One patient experienced a 19% decrease in FEV1 following HTS and required rescue albuterol treatment, after which the FEV1 returned to baseline. The pre-HTS mean oxyhemoglobin saturation on room air was 95%. Two patients had brief desaturation (87% and 88%), which resolved without intervention. Participants reported increased cough (82%), wheezing (27%) and chest tightness (18%) with HTS. Three patients who experienced hemoptysis had prior history of similar symptoms at baseline during airway clearance.

Initiation of hypertonic saline under close observation in CF patients with severe airflow obstruction seems to be generally safe and well tolerated, though considerable individual variation was observed. A larger, prospective, controlled study would be useful to assess the long-term safety and efficacy of HTS treatment in this patient population. Background: Cystic fibrosis (CF) leads to a progressive decline in lung function and exercise tolerance. Because regular exercise has the potential to improve pulmonary function, developing a routine exercise program may be beneficial for patients with CF. Pulmonary exacerbations are the leading cause of morbidity among CF patients, but it is unknown how pulmonary exacerbations affect exercise tolerance. Therefore, we investigated the exercise tolerance of children with CF hospitalized for a pulmonary exacerbation using the Modified Shuttle Walk Test (MSWT).

Methods: Exercise tolerance of pediatric CF inpatients was assessed using the MSWT, a reliable measure of exercise tolerance in which the subject walks 10m lengths (shuttles) at progressively increasing paces. Pulmonary function testing was performed in clinic on the day of admission and the MSWT by the inpatient physical therapist within the first few days of hospitalization. Along with measurement of oxygen saturation and heart rate, patients completed the Modified Borg Dyspnea Scale (MBS) and the Modified Borg Rating of Perceived Exertion (RPE) to estimate the degree of breathlessness and exertion on a scale from 0 (none) to 10 (maximal) immediately before and after the MSWT.

Results: Twenty-five inpatients (3M, 22F) performed the MSWT, five of whom performed the test at both admission and discharge. Median patient age was 14.8y (range 7 -19y). Median FEV 1 (% predicted) was 51% (range 20-98%). Patients completed a median of 59.5 shuttles (range 20-136) during the MSWT, with a median percent change in heart rate of 31.3% (range -0.8-82.2%). The median MBS value increased from 0 (range 0-4 pre-MSWT) to 6.5 (range 0-10 post-MSWT). The median RPE value was 0 (range 0-5 pre-MSWT) and increased to 4 (range 0-10 post-MSWT). The five patients who performed the MSWT at admission and discharge all completed more shuttles on the second test, with a median percent increase of 10.6% (range 2.9-25.0%). FEV 1 correlated with the number of completed shuttles (r 2 =0.45) and approached significance (p=0.053), demonstrating that patients with higher lung function perform better on the MSWT.

Conclusions: Inpatients undergoing treatment for pulmonary exacerbations reported severe breathlessness and moderate exertion following exercise using the modified Borg scales. These results demonstrate that during a pulmonary exacerbation, patients are limited by dyspnea and those with lower lung function have lower exercise tolerance. Furthermore, exercise tolerance improves over the course of relatively short hospitalizations and also may improve as a consequence of patient motivation to learn the test and perform it better. These findings show that MSWT is a useful measure of improvement in the hospital and has potential applications to assess exercise tolerance in an outpatient setting.

Pareek, N.; George, P.; Cowman, S.; Gyi, K.; Mullen, M.

Introduction: The prevalence of patent foramen ovales (PFOs) in cystic fibrosis (CF) patients is unknown. Transient ischaemic attacks (TIAs) have been reported in CF patients with PFOs. PFOs are also implicated in respiratory disease as an aetiology of hypoxia and reduced performance status. We report 4 cases where PFOs have been identified in CF patients. We identify whether PFOS play a possible role in hypoxaemia in these patients and the benefits and safety of percutaneous closure.

Methods: Between 1999 and 2008, the CF Registry at the Royal Brompton Hospital was reviewed to identify CF patients with PFO with subsequent closure. PFOs were diagnosed by trans-thoracic echocardiography using contrast injection. Four patients were identified. Data were recorded from their hospital records and was analysed, assessing presentations, echocardiographic findings and subsequent management.

Results: Four patients were identified. Three were female and 1 was male. Median age at clinical presentation was 25 (range 18 -33). Median follow-up was 5 years from initial presentation to either current date or death (range 10 months to 12 years.) Median FEV 1 at presentation of symptoms was 1.64L/min (range 0.9-1.8) with median FEV 1 predicted 48.4% (range 30%-60%).

Patients A and B presented with a TIA, Patient C presented with a TIA and persistent hypoxia and Patient D presented with neurological symptoms during physiotherapy. Three patients had implantable venous access devices (IVADs) in situ; these patients all presented with TIAs. Patient A was also lupus anticoagulant positive and was commenced on lifelong warfarin.

All patents were closed with an Amplatzer PFO occluder device. The procedures were uncomplicated. Major adverse cardiac events were zero at 30 days, 1 year and at follow-up. Table 1 shows the effect of PFO closure on oxygenation in these patients. One patient (Patient A) had TIAs post-closure with an episode of quadrantanopia which persisted. She was placed on the transplant list but died 2 years post-closure of PFO due to an episode of massive haemoptysis. The remaining 3 patients had no further neurological episodes and are well at follow-up.

Conclusions: CF patients may be at increased risk of paradoxical embolisation compared with the general population. The presence of IVADs, physiotherapy, coughing paroxysms and obstructive lung disease increase the risk of right to left shunting of thrombus. It is safe and potentially beneficial to close PFOs in CF patients, with a reduced risk of neurological episodes and possible improvement in oxygenation. No improvement in lung function was seen. PFOs should be considered as an important aetiology of neurological complications and deteriorating oxygenation in CF patients. Table 1 . Partial pressure of oxygen (kPA) and oxygen saturations (in brackets) related to PFO-closure. Background: Aquagenic wrinkling of the palms (AWP) is a cutaneous phenomenon marked by the transient formation of edematous, translucent papules and plaques on the palms and fingertips within minutes of exposure to water. In the few reports in the literature, it appears that AWP may be associated with cystic fibrosis (CF). We hypothesized that AWP is an underreported cutaneous manifestation of CF. The primary aim of the trial was to establish the prevalence of aquagenic wrinkling in children with cystic fibrosis relative to controls. <\>> Methods: Fifty children with CF and 25 control children without CF who were being treated for asthma underwent a 5 minute hand immersion in lukewarm water to attempt to induce AWP. The test was considered positive if subjects demonstrated >30% wrinkling over the palmar surface area. Among subjects with CF, secondary correlations were attempted to deter-mine whether CFTR genotype/mutation class, pancreatic sufficiency, body mass index (BMI), sweat chloride levels, palmar transepidermal water loss as measured by a vapometer, and most recent FEV1% predicted were associated with the presence or absence of AWP.

<\>> Results: Forty (80%) of the subjects with CF demonstrated aquagenic wrinkling. In contrast, none (0%) of the controls had aquagenic wrinkling (p<0.001). These results remained statistically significant when stratified for by age and race, to account for differences in the baseline characteristics of the 2 groups. Transepidermal water loss, as measured by the vapometer, was significantly higher in CF subjects with AWP compared to those without (p<0.001). There was no correlation of the presence versus absence of AWP in subjects with CF and either genotyope/CFTR mutation class, pancreatic sufficiency, sweat chloride, FEV1 % predicted or BMI, although these analyses were likely underpowered.

<\>> Conclusions: Aquagenic wrinkling of the palms is common in children with CF, and is associated with increased transepidermal water loss. Further study is necessary to determine whether aquagenic wrinkling is a characteristic and discriminating feature of CF, or whether it is associated with CFTR genotype, sweat chloride or other clinical parameters. Additional study is also necessary to determine if AWP may be a useful physiologic outcome measure for studies of CFTR potentiators and/or correctors. Rationale: Most patients with cystic fibrosis (CF) are detected through neonatal screening or by symptoms. Lung disease in CF is at the beginning a disease mainly of the small airways. Although infants with CF have normal lungs at birth, respiratory dysfunction often develops in the first year.

Objective: To assess airway function in children younger than 24 months detected through neonatal screening and by symptoms in follow-up at our center.

Methods: Cross sectional study. We compared two groups of patients with similar age: group A detected by neonatal screening (GA), and group B by symptoms (GB), both diagnosed by sweat test. For neonatal screening we used immunoreactive trypsinogen (IRT) in dried blood spots (cut-off value: 70 µg/ml). We measured Maximal Expiratory Flow at FRC (V'maxFRC) in partial expiratory flow-volume curves (PEFVCs) at the point of functional residual capacity (FRC). PEFVCs were obtained using the rapid thoracoabdominal compression technique with Jaeger ® equipment. Student's t-test was performed to compare the means of the indicators using the SPSS software 9.0.

Results: Twenty infants with CF were enrolled (8 in GA and 12 in GB) with a mean age of 15.9 ± 8 months, height 73 ± 9 cm, z score for weight/ age -1.2 and height/age -1.3. Fourteen patients finished all measurements (6 GA and 8 GB). Group A showed higher values of V'maxFRC than group B, as shown in the table (p< 0.05).

Conclusion: Our infants with cystic fibrosis detected through neonatal screening had higher V'maxFRC than the non-screened group. These findings should be considered in the care and treatment indication as well as when CF programs are considered. Lebanon, NH, USA; 2. Center for Environmental Health Sciences, Dartmouth College, Hanover, NH, USA; 3. Physiology, Dartmouth Medical School, Hanover, NH, USA; 4. Microbiology and Immunology, Dartmouth Medical School, Hanover, NH, USA A body of evidence now cites a synergistic relationship between Pseudomonas aeruginosa (PA) airway colonization and bioavailable iron in airway secretions. Sputum from cystic fibrosis (CF) patients contains more iron than that obtained from healthy controls, and PA growth from CF sputum is directly proportional to sputum iron and ferritin concentrations [1] .

The present study was conceived to elucidate novel relationships among hematologic and sputum iron indices and clinically-relevant metrics of CF care. Preliminary data from fifteen stable outpatients and eight patients hospitalized for exacerbation were analyzed. Spearman's rank correlation (ρ) was used to determine relationships among biochemical and categorical variables (Table 1) . Lower serum iron levels were associated with inpatient status, diabetes mellitus, inhaled tobramycin and dornase alfa use, and lower FEV1. These data suggest that iron-deficiency anemia is a surrogate for disease severity in cystic fibrosis. FEV1 was inversely related to serum interleukin-6 (IL-6) (R 2 =0.25, p=0.03) and erythropoietin levels (R 2 =0.56, p <0.001) but was directly related to serum iron level (R 2 =0.34, p=0.006). Iron-deficiency also correlated with higher serum IL-6 levels (ρ = -0.55, p=0.01). Despite its reported accuracy in differentiating anemia of chronic inflammation from iron-deficiency anemia, [2] the soluble transferrin receptor (sTfR) level correlated only with serum erythropoietin (R 2 =0.31, p=0.009). Quantification of sputum iron levels for the cohort is forthcoming. This is the first study in cystic fibrosis to explicitly associate lung function and systemic iron deficiency with circulating IL-6. IL-6 causes hypoferritinemia by inducing the production of hepcidin, a regulator of enterocyte iron uptake. [3] The role of hepcidin in the anemia of cystic fibrosis, however, is unclear. Because sTfR levels did not correlate with any hematologic iron parameter, this test may have limited clinical utility in working up anemia among CF patients. Background: There are currently no data in adult cystic fibrosis (CF) patients on the prevalence of hyperinflation and its relation with dyspnea, quality of life and exercise performance in this population. Furthermore, little is known about the functional correlates of hyperinflation in CF.

Methods: We evaluated 84 CF patients (mean age: 29 [16; 70] ). Lung function tests included flow volume curve, plethysmography and exercise capacity during a 6-minute walk test. Hyperinflation was defined by a functional residual capacity (FRC) > 120 % pred. In order to evaluate the involvement of small airways in the development of hyperinflation, airway resistance were measured at various frequencies by the Forced Oscillation Technique (FOT).

Correlations between hyperinflation, dyspnea (MRC), exercise performance and quality of life with a disease-specific health-related QoL questionnaire (CFQ 14+) were also assessed.

Results: A significant hyperinflation was present in 53.5% of patients. Using the inspiratory capacity to total lung capacity ratio (IC/TLC) as an index of hyperinflation severity, we found a strong correlation between IC/TLC and MRC scale (r = 0.52 ; p = 0.01), distance walked in 6 minutes (r =0.5 ; p<0.01), the respiratory symptoms (r=0.42 ; p<0,01), physical functioning (r=0.6 ; p =0.001) and health perception scale (r=0.52 ; p<0.01) dimensions of the CFQ14+.

FRC and Rrs at 0Hz (evaluating small airways) were importantly correlated (r=0.63; p<0.001), as Raw and Rrs 0Hz (r=0.82 ; p<0.001), while there were no correlations between FRC and Rrs 16Hz or 24 Hz (reflecting central airways).

Conclusion: Hyperinflation is frequent and under-evaluated in adult cystic fibrosis. It could be partly due to the small airway disease as suggested by low frequency Rrs. The fraction IC/TLC seems a relevant parameter to predict dyspnea, exercise performance and lower quality of life. 

When assessing a patient with cystic fibrosis (CF) for transplantation, FEV 1 is one of the key factors that is considered. Kerem et al. (N Engl J Med 1992; 326:1187) demonstrated that once the FEV 1 of a patient with CF drops below 30% of the predicted value, the mean mortality is greater than 50% within 2 years. Our clinical experience has led us to believe that with modern treatments, this may no longer be the case. We therefore wish to test our hypothesis that patients with this level of lung function now survive beyond the previously quoted 2 years.

Utilising The Royal Brompton CF database, we identified all patients that had reached an FEV 1 <30% of the predicted value between 1990 and 2007. The date at which patients first recorded this level of lung function was their date of entry into the study and these patients were subsequently followed up until their date of death or the end of the study.

The data set (n=395) was examined using survival analysis methods. Two groups were analysed -Group 1 (n=207) entering the study between 01/01/1990 and 31/12/1996 and Group 2 (n=188) entering the study between 01/01/1997 and 31/12/2007. Across the entire cohort, the range of ages was 14.2 to 67.6 years old. The mean age in Group 1 was 26.2 years old and in Group 2 was 26.0 years old. Using Kaplan-Meier survival analyses to compare the two groups, the median survival in Group 1 was 2.1 years compared to 5.4 years in Group 2 and this difference was statistically significant (p<0.0001). Both groups were statistically similar with regard to mean age, gender, prevalence of diabetes, long term oxygen therapy, use of supplementary pancreatic enzymes and lung function at entry into the study. The only difference between the two groups was the use of rhDNase. Use of rhDNase was much greater in the group with improved survival. Significantly more patients in Group 2 took rhDNase regularly with 35% in Group 1 vs. 87% in Group 2 (p<0.001) and using Cox regression analyses, patients who were using rhDNase had a significantly reduced risk of mortality by 41% (p<0.001). Diabetes conferred a significantly increased risk of death by 74% (p<0.001) and female gender by 38% (p<0.01) across both groups.

This data shows that the survival of patients with CF at low levels of lung function has significantly improved. We have in fact underestimated this improvement as some in Group 2 were in the study for a shorter period of time than the patients in Group 1. Many factors may be implicated but it would appear that the regular use of rhDNase, which became widespread in the mid 1990s, is associated with stabilisation of lung function. Further analyses need to be undertaken but it is no longer valid for clinicians to assume that an FEV 1 <30% of the predicted value means that 50% of patients will die within 2 years. Consideration should be given to treatment with rhDNase if this is not already the case. In addition, it is important to note that female patients and those with diabetes might be considered for transplantation at an earlier stage. . Adults with CF are living longer with more complex disease, requiring novel approaches to treatment. Although standard airway clearance techniques (e.g. ACBT, autogenic drainage, oscillating PEP devices) are the foundation of independent daily management, physiotherapists are forced to widen their treatment repertoire as advanced disease challenges our ability to achieve effective airway clearance. The use of mechanical adjuncts in combination with our standard techniques may help to overcome the difficulties faced.

Aim: To illustrate the use of mechanical adjuncts in adults with infective exacerbations of CF.

Method: A retrospective analysis of the physiotherapy and clinical records of 19 adults with CF admitted with an infective exacerbation (Jan 2008-Apr 2009), treated using a mechanical adjunct for airway clearance. These devices consisted of High Frequency Chest Wall Oscillation (HFCWO) (The Vest ® ), Intrapulmonary Percussive Ventilation (IPV) (IMPII ® ), and Mechanical In-Exsufflation (MI-E) (Cough-assist ® ). Data analysed: clinical reasoning for initiation; duration of use; patient outcome.

Results: Lung function at initiation: Median FEV1 0.77L (range 0.22-2.12), median FVC 1.87L (range 0.79-3.72). Length of admission 15-273 days where 16% of patients were intubated, and 53% used non-invasive ventilation. Medical management included multiple IV antibiotics (100%), hydrocortisone (89%), aminophylline (68%), and 47% of patients had their DNase dose doubled. Over a 16 month period 19 patients used a mechanical adjunct in addition to and in combination with their usual airway clearance technique: 58% (11/19) used HFCWO; 21% (4/19) used IPV; 10.5% (2/19) used a combination of IPV with MI-E; 10.5% (2/19) used MI-E. Duration of use of a device ranged from 1-90 days. The primary and secondary reasons for initiation of the adjunct were analysed for each device (Table 1) .

Discussion: The study group represents a population of adults with advanced and complex disease, eight of which were in their final admission. This necessitated the use of treatment combinations using mechanical adjuncts to optimise airway clearance, prompted by a range of clinical observations. In advanced disease and respiratory failure where symptoms such as fatigue and airway tightness compound active airway clearance techniques, mechanical adjuncts may assist in conserving energy whilst maintaining airway clearance. In patients who are unable to fully participate in physiotherapy and/or continue to deteriorate, mechanical adjuncts offer the ability to continue an airway clearance regime, and the benefits of these devices are worthy of further investigation in a prospective study. 1 ; Gunaratram, C. 2 1. School of Health and Human Performance, Dublin City university, Dublin, Ireland; 2. Respiratory Department, Beaumont Hospital, Dublin, Ireland Background: Peak exercise performance is reduced in cystic fibrosis (CF). An inefficient breathing pattern contributes to the reduction in peak exercise performance in CF. The ability to achieve a high tidal volume (VT) at peak exercise is limited in CF, and is more pronounced as disease severity increases. The purpose of the study was to compare breathing pattern at peak exercise in a wide range of CF disease severities.

Methods: A total of 29 CF patients (24±5 yr) with different severities of CF and 15 healthy age matched controls (24±4 yr) performed a maximal exercise test on a cycle ergometry using an incremental protocol. The FEV1 % predicted was 86±12, 50±8, and 28±2.0 for patients in the mild (n = 16, f = 3), moderate (n = 7, f = 2), and severe (n = 6, f =1) CF categories, respectively.

Results: Oxygen uptake (VO 2 ) and workload at peak exercise were (i) higher in healthy controls than CF patients; (ii) similar in patients with mild and moderate CF; and (iii) lower in severe than mild and moderate CF. VT was similar in healthy controls and mild CF, and significantly different between severe CF and the other groups. There was a relation between VT max and peakVO 2 corrected for lean body mass (r = 0.71 P< 0.001) and between VT max, and peak lactate (r = 0.59 p< 0.01).

Conclusion: The ability to increase VT at maximal exercise is limited mainly in severe CF. By increasing VT, patients with moderate CF are able to adopt a breathing pattern that is very similar to mild CF and therefore can tolerate high peak lactate and achieve a peak maximal performance similar to mild CF. Regular training may be very beneficial in mild and moderate CF. Patients with severe CF may require specific strategies that may help to increase VT during exercise. By increasing VT, severe patients may be able to increase exercise intensity that will be enough to yield a physiological training effect.

The study was supported with a grant from CF HopeSource Ireland. 

Doumit, M. 1 ; Krishnan, U. 3 ; Jaffé, A. 2 ; Belessis, Y. 2 1. Physiotherapy, Sydney Children's Hospital, Sydney, NSW, Australia; 2. Respiratory Medicine, Sydney Children's Hospital, Sydney, NSW, Australia; 3. Gastroenterology, Sydney Children's Hospital, Sydney, NSW, Australia Background: There is conflicting evidence regarding the effect of the head-down position on gastro-oesophageal reflux (GOR) during physiotherapy in young children with cystic fibrosis (CF). Furthermore there is no evidence on the impact of physiotherapy on GOR as assessed by pH-impedance monitoring, which detects acid and non-acid reflux and its proximal extent.

Aims: 1) To determine whether positioning young children with CF in the head-down position during chest physiotherapy is associated with worsening of GOR as determined by pH-impedance monitoring. 2) To quantify acid and non-acid GOR in young children with CF and determine whether GOR extends sufficiently proximally to potentially directly impact on the airways.

Methods: Subjects were studied using pH-impedance monitoring over 24 hours. During this period, subjects received two 20-minute periods of chest physiotherapy in random order. One session comprised percussion and vibration in "modified" postural drainage positions with no head-down tilt and the other session comprised percussion and vibration in gravityassisted postural drainage positions, which included head-down tilt. Our primary outcome measure was number of reflux episodes detected by impedance during physiotherapy. It was determined that 16 subjects would be needed to have an 80% chance of showing a two-fold increase in number of reflux episodes in the head down positions (α = 0.05).

Results: Seventeen young children with CF (6 males), median age 15 months (range 8-34) participated. When the two physiotherapy sessions were compared, there was no significant difference in number of reflux episodes detected by impedance or any other reflux parameter measured as summarised in the table below. The results of the 24 hour investigations revealed 53% of subjects had abnormal acid reflux index (RI) for age as determined by pH monitoring (median 7.3% range 0.5-17.6%). Using 24 hour impedance data, the median total (RI) was 2.95% (range 0.6-9.6%), median acid RI was 2.08% (range 0.8-2.4%) and median non-acid RI was 0.5% (range 0-5.9%). Combining data from all subjects, a total of 966 reflux episodes were detected by impedance, 658 (68%) were acid and 308 (32%) were non-acid. Importantly, in every child, a high proportion of reflux episodes reached the proximal oesophagus with a median of 71.5% (range 55-85%) of reflux episodes reaching proximally.

Conclusion: Physiotherapy in the head-down position does not appear to exacerbate acid or non-acid GOR as assessed by impedance. Using the pH-impedance technique, we have shown that GOR with proximal extension is common in young children with CF.

Research kindly supported by Sydney Children's Hospital Foundation.

Values expressed as median (IQR . At our Pediatric CF Center, we began prescribing overnight use of this device to some of our sicker patients in an attempt to change the course of their disease. As it became apparent that overnight use was tolerated by these patients, we started recommending it to patients who found it difficult to adhere to daytime HFCWO, usually because of time constraints, and later to other patients. This report describes the receptiveness of our patients to overnight HFCWO and data about our initial experience.

Methods: A chart review was conducted of the 135 pediatric patients at our Center to identify all patients with whom overnight HFCWO was discussed. For those who used this therapy for more than 6 months, we recorded the best forced expiratory volume in one second (FEV1) percent predicted during the baseline 6 months prior to initiation of overnight therapy, and the best FEV1 during the 6 months immediately thereafter. Patient reports regarding tolerability of overnight therapy were compiled.

Results: Through February 2009, overnight HFCWO was discussed with 43 patients. An upgrade from an older device was required for all interested patients. Thirty patients agreed to investigate whether they were able to fall asleep while using HFCWO, and 22 reported they were able to do so. Participation was limited by insurance restrictions. Of 13 patients who started overnight therapy none discontinued it. The following data is from 9 patients who completed at least 6 months of therapy. Their mean age at initiation was 11 years (range, 7-19); 5/9 were male; and mean baseline FEV1 was 88% of predicted (range, 44-109%). Patients used overnight HFCWO for a mean 3.5 hours/night (range, 1-8 hours), and 3/9 patients continued twice daily daytime HFCWO as well. As compared to baseline, FEV1 improved in 7/9 patients, and the mean change in FEV1 was +6% (range, -21% to +29%). The 19-year-old whose FEV1 fell by 21% was found to be newly colonized with Pseudomonas aeruginosa. A 13-year-old whose FEV1 was 44% at baseline had a very productive cough, and was receiving nocturnal feedings via gastrostomy tube. She awakened twice a night to expectorate, and immediately fell back asleep. She did not experience any emesis or difficulty with digestion. Her FEV1 improvement was 27%. Two patients complained the vest used to deliver HFCWO sometimes became too hot to use. Four patients found it easier to use a wrap vest as compared to the full or chest vests.

Discussion: These preliminary findings demonstrate that patients with CF can tolerate overnight HFCWO. As this is an uncontrolled report, it cannot be concluded on the basis of this data whether overnight HFCWO improves pulmonary function as compared to daytime therapy. The data provide assurance and impetus for further research regarding this therapeutic modality.

Collins, N. 1 ; Gupta, A. 2 ; Bush, A. 2 ; Urquhart, D. 2 1. Physiotherapy, Royal Brompton and Harefield NHS Trust, London, United Kingdom; 2. Paediatric Respiratory Medicine, Royal Brompton and Harefield NHS Trust & Imperial College, London, United Kingdom Introduction: With improvements in clinical care, children with CF are tending to be healthier and less likely to develop respiratory failure until later on in life. A recent randomised, crossover study in adults with CF has suggested improved daytime functioning following nocturnal non-invasive ventilation (NIV) (1) . There is much less experience in children and its use in the paediatric setting has not been truly established.

Aim: To determine the extent of NIV use in paediatric CF within the UK, to ascertain if practice is similar between centres, and to see what leads to initiation of NIV in CF children.

Method: A semi structured questionnaire was sent to the CF consultant and physiotherapist of the 27 paediatric specialist CF centres in the UK.

Results: Twenty of 27 centres responded (74%) representing 3149 children. Only 8 patients were on NIV (0.25%), median age 14 years (range 10 to 16 years) at the time of initiation. The mean usage of NIV was 7.14 hours per day. Forty-five percent of the centres (9/20) undertook regular sleep studies, the frequency of these varying between 1 per week (4 centres) to 1 per year. Fifty-five percent used overnight oximetry, 25% transcutaneous pCO2 and 5% full polysomnography. Another 2 centres took early morning capillary gases if the patients FEV 1 was <50% predicted. Hypoxia was variably defined as SpO 2 <93% to <90% for >10% of the study, and hypercapnia variably between pCO 2 > 6.5 kPa to pCO 2 >8 kPa. NIV was initiated as a bridge to transplant, as an adjunct to physiotherapy or in an acute exacerbation. BiPAP was used in 20 centres (75%), single level ventilation 20% and volume control ventilation 5%. NIV was most commonly set up by the physiotherapist; only 6/20 centres had a protocol for NIV initiation. There were 10 reported NIV failures; claustrophobia and discomfort were the main reasons.

Conclusion: NIV is rarely used in UK CF children. There is heterogeneity both to assessment of respiratory failure and use of NIV. There is no agreed definition of hypoxia and hypercapnia, modes of NIV used differs, and few institutions have a protocol for initiation of NIV. Guidelines for NIV use in children with CF are needed.

1. Young AC, Wilson JW, Kotsimbos TC, Naughton MT. Randomised placebo controlled trial of non-invasive ventilation for hypercapnia in cystic fibrosis. Thorax 2008;63:72-77.

MacFadyen, K. 1,2 ; Beairsto, C. 2 ; Mercer, B. 2 ; Parsons, J. 2 ; Trecartin, L. 2 ; Dechman, G. 2 1. Physiotherapy, IWk Health Centre, Halifax, NS, Canada; 2. Physiotherapy, IWK Health Centre, Halifax, NS, Canada Introduction: Limited aerobic capacity in children with cystic fibrosis (CF) may affect their physical functioning and quality of life (1) . There are several valid and reliable field tests for measuring aerobic capacity in children with CF, but there are no such field tests to measure aerobic endurance. In adults with chronic obstructive pulmonary disease (COPD), endurance has been shown to be a stronger indicator of function than aerobic capacity (2) . The Endurance Shuttle Walk Test (ESWT) is reliable and valid in adults with COPD (3).

Purpose: The purpose of this study was to test the reliability of the ESWT in children with CF.

Methods: Six patients (3 male/3 female) attending the CF Clinic at the IWK Health Centre (average age 11.3, ±2.6 years) were recruited to the study. The subjects completed the Modified Shuttle Walk Test (MSWT) as part of their clinic visit. The MSWT was used to predict each subject's VO 2 max (4). A value equivalent to 85% of the subject's VO 2 max was calculated and the walking speed for the ESWT was interpolated from a graph by Revill et al. (4) . Three ESWTs were performed and assessed for reliability. During each test the following outcome measures were collected: respiratory rate (RR), heart rate (HR), oxygen saturation (SpO 2 ), systolic and diastolic blood pressure (SBP and DBP), modified Borg rating of breathlessness (BORG rating), leg fatigue (LF), distance covered, and endurance time.

Data Analysis: A 2-way analysis of variance (ANOVA) was used to assess differences in change in HR among the three trials. Intraclass correlation coefficients (ICC) were used to describe the reliability of the measurements.

Results: There was no statistically significant difference in resting HR or HR change over the 3 trials. There was also no statistically significant difference in post-trial HR, RR, SpO 2 , SBP, DBP, BORG ratings, and LF. The ICC for the distance covered in the second and third ESWT showed moderate reliability (r = 0.799).

Discussion: The results of this study demonstrate that the ESWT is moderately reliable in this small cohort of children with CF (5). Reliability was not affected by a subject's baseline physiological status as there was no significant difference in pre-test physiological parameters. Reliability may have been affected by differences in subject motivation across the trials and learning effects among trials.

Conclusion: The ESWT is moderately reliable in children with CF. Further research in this area with a larger number of subjects is warranted to confirm these results. Several studies have shown an inferior exercise performance in patients with cystic fibrosis (CF) compared with healthy individuals when looking at the data in absolute terms or expressing the measurements relative to body weight. Furthermore, one study employing 31 P-magnetic resonance spectroscopy (MRS) suggested a disturbed muscle metabolism during isometric quadriceps knee extension exercise to exhaustion in CF. Based on these data, an involvement of skeletal muscle in CF has been hypothesized. The objectives of this study were 1) to assess differences in exercise performance between patients with CF and a control group while using allometric scaling to account for differences in body size, and 2) to assess muscle metabolism by 31 P-MRS acquired during dynamic exercise of the quadriceps femoris muscle.

Sixteen patients with CF (5 female; FEV1 75±25%, range 25-113%) and 16 gender and age matched controls (FEV1 108±10%, range 93-125%) aged 12-42 years participated in this study. Muscle power (MP) was assessed using the Wingate test and peak oxygen uptake (VO 2 peak) was determined during incremental cycle ergometry (Godfrey protocol). Quadriceps muscle cross sectional area (CSAquad) was determined by magnetic resonance imaging at mid thigh while muscle phosphocreatine (PCr) and inorganic phosphate (Pi) were assessed by 31 P-MRS at rest and during maximal dynamic leg extension exercise.

Patients with CF were significantly smaller (163±11 vs. 176±9 cm) and lighter (56±13 vs. 70±9 kg) than controls. They also had a lower CSAquad (57±14 vs. 69±11 cm 2 ). MP (369±120 vs. 519±126 W), VO 2 peak (2045±522 vs. 2929±764 ml/min) and maximal load during leg extension exercise (6.7±1.9 vs. 8.3±1.6 kg) were also significantly lower in CF. However, the differences in performance between groups vanished when values were adjusted by allometric scaling for CSAquad and, for MP and maximal leg extension work rate only, height. There was no difference between CF and non-CF subjects at rest and during maximal leg extension exercise in the area under the curve for the PCr and Pi peaks in the 31 P-MRS spectra nor in the quotient Pi/PCr. Furthermore, there was no difference between groups in absolute or relative changes of these parameters from resting values.

In conclusion, differences in muscle performance between patients with CF and healthy controls could be explained by differences in body size. There was no difference in muscle metabolism between groups at rest or during dynamic exercise. These data do not support the hypothesis of a primary involvement of skeletal muscle in CF.

This study was supported by a grant from the German CF foundation Mukoviszidose e.V.

Button, B. 1,2 ; Rasekaba, T. 3 ; Wilson, J. 1, 2 ; Holland, A.E. 1, 3 1. AIRmed, The Alfred Hospital, Melbourne, VIC, Australia; 2. Medicine, Monash University, Melbourne, VIC, Australia; 3. Physiotherapy, La Trobe University, Melbourne, VIC, Australia The three-minute step test (3MST) is a simple test of exercise capacity which is commonly used in patients with cystic fibrosis (CF), however there are no data regarding its prognostic value. This study aims to determine whether performance on the 3MST is associated with long-term clinical outcomes in adults with CF.

Consecutive patients in a stable state of health were recruited from the CF outpatient clinic at The Alfred Hospital. The test was conducted following a standardised protocol using a 15cm high step, external pacing at 30 steps/minute, continuous monitoring of SpO 2 , heart rate and perceived exertion on the Borg scale. Test variables included changes in heart rate, dyspnoea, perceived exertion, SpO 2 and difficulty rating on the visual analogue scale. Relationships between baseline performance on the 3MST and clinical outcomes at 12 months (change in % predicted FEV1 and inpatient hospital days) were assessed.

Results: Data were available for 48 (30 males) of mean age 30 (standard deviation 8) years, FEV1 64 (25) % predicted and BMI 23 (5) kg.m 2 . Participants who desaturated to less than 90% on 3MST (n=12) had a larger number of hospital days over the following 12 months than those who did not (median 17 days vs 2 days, p= 0.007). Those who desaturated also had a greater decline in FEV1% predicted (mean difference 4.9%, 95% confidence interval 0.5 -9.3%). Heart rate and symptom scores on 3MST were not associated with clinical outcome at 12 months.

Conclusions: Preliminary results indicate that performance on the 3MST may be associated with longer-term pulmonary deterioration and increased hospital admission days. Larger samples are required to confirm the prognostic value of the 3MST in adults with CF. Introduction: Exercise testing has become an important tool in the evaluation and care of patients with cystic fibrosis (CF). Exercise capacity decreases with progression of lung disease, and in severely affected patients with CF supplemental oxygen during exercise, or even for 24 hours may be warranted to avoid oxygen desaturation. In those patients with CF having supplemental oxygen a cardio-pulmonary exercise test (CPX) is difficult to perform. Due to various factors (e.g. deconditioning, peripheral muscle dysfunction) the results of CPX may not be useful to determine training intensity in most of the patients. The aim of the present study was to use a stepramp test (SRT) on a treadmill to determine training intensity and to evaluate a training program for patients with advanced lung disease.

Material/Method: We studied 20 patients with CF aged 16 to 41 years (age 26.4±7.5yrs, FEV1 38.0±10.9%pred). All patients performed a SRT on a treadmill using a modified Balke protocol. The velocity was continuously 4 km/h during the test. The initial grade was 2.5%, with an increment of 2.5% every 20 sec. The test was stopped when the patients were exhausted. During the test, heart rate (HR) and oxygen saturation (O 2 Sat) were recorded. All patients received supplemental oxygen during the STR. After 6-week training program the SRT was repeated. The training program consisted of an interval-training on a treadmill (continuously 4 km/h) with duration of 20 minutes five times/weekly. High intervals of 30 s with 50% of maximal grade were followed by an active recovery of 0% grade. In total 10 periods of high intensity were followed by active recovery. Supplemental oxygen was administered to reach a haemoglobin oxygen saturation of more than 90% during training. Discussion: Most SRT for patients with CF are performed on an electronically braked bicycle. We adapted the SRT to a treadmill. The test was well tolerated by all patients. Most of the patients stopped the test due to muscle fatigue before as well after training. Recommendation of training intensity could be given according to the results of the SRT (50% of max grade) achieved during SRT. A significant training effect was observed after the exercise program of 6-weeks.

The SRT is easy to administer and takes relatively little time. The SRT could be repeated at short intervals (weekly or every two weeks) to adopt training intensity. Muscle fatigue was the most important factor when test was stopped and not dyspnoea. Thus the SRT provides one way to determine training intensity in patients with advanced lung disease and to investigate effects of an exercise training program.

Wu, K. 1, 2 ; Fox, P. 3, 2 ; Tullis, E. 1, 2 ;  Stephenson, A.L. 1, 2 1. Adult Cystic Fibrosis Program, St. Michael's Hospital, Toronto, ON, Canada; 2. University of Toronto, Toronto, ON, Canada; 3. Mobility Program Clinical Research Unit, Keenan Research Centre, Li Ka Shing Knowledge Institute, St. Michael's Hospital, Toronto, ON, Canada Introduction: Chest physiotherapy (CPT) is vital in facilitating secretion clearance from the airways in cystic fibrosis (CF). Nonadherence to CPT is however well recognized in the literature. To enhance patients' intention to do regular CPT, we produced a video entitled "Breathe Easier: Airway Clearance Techniques (ACT) for People with Cystic Fibrosis." This study evaluates the effectiveness of the video.

Methods: Sixty individuals attending the Adult CF Clinic at St. Michael's Hospital were recruited for this study. The video was evaluated using the theory of planned behaviour (TPB) framework. The theory asserts that individuals have a greater behavioural intention and are more likely to perform behaviours if they evaluate a behaviour as positive (attitude towards behaviour), they think their significant others want them to perform the behaviour (subjective norm), and they think the behaviour is easy to perform (perceived behavioural control). Behavioural intention, defined as how strongly patients intend to do regular CPT in the next three months, was the primary outcome measure. Secondary outcome measures included patients' behavioural beliefs, attitude towards behaviour, subjective norms, and perceived behavioural control. Patients were asked to complete a questionnaire, view the video, and complete a post-video questionnaire. The questionnaires were developed based on the TPB framework. Data were collected on patient demographics, FEV 1 , self-perceived health, and CPT regimen. Descriptive statistics (mean±SD) were carried out. A paired t-test was used to compare responses before and after viewing the video.

Results: Preliminary data analysis was available on 34 patients who have completed the study to date. The mean age was 32±11y; FEV 1 was 54±20%pred. Seventy-six percent of subjects were male. Prior to viewing the video, 76% of patients reported doing regular CPT; however, only 44% did CPT as recommended. After watching the video, patients' behavioural intentions to do regular CPT significantly increased (p=.002). Significant increases were also found in patients' attitudes towards behaviour (p=.0002), behavioural beliefs (p=.005), and perceived behavioural control (p=.003) after watching the video. No significant change was noted in patients' subjective norms. Ninety-one percent of patients reported hearing about and/or seeing examples of ACT for the first time from the video. Watching the video stimulated questions regarding CPT in 38% of patients. Among those who did not do regular CPT, 50% had questions for the physiotherapist after watching the video.

Conclusions: The preliminary analysis on 50% of the anticipated sample size showed that viewing the video improved patients' intention to do regular CPT. Improvements were seen in patients' attitude towards doing CPT, their belief that positive outcomes will result from doing CPT, and their perception of their capability to do CPT after watching the video. The video stimulated questions regarding CPT which may result in better adherence to CPT. Further analysis with the full cohort is needed before firm conclusions can be drawn. Cystic fibrosis (CF) is characterised by recurrent bacterial infections of the lungs which eventually cause irreversible lung damage. Inhaled antibiotic therapy in the home environment may postpone or prevent hospital admissions. The increasing use of inhaled antibiotics in combination with inhaled mucolytics, can create a time burden that is often underestimated by those involved in CF care. Furthermore, there is often confusion about the order and timing of inhaled medications around other aspects of CF care.

To gauge an overview of nebuliser use and compliance within our paediatric clinic we developed a prospective survey that addressed all aspects of use and care of nebulisers. Surveys were completed during clinic visits by 75 families with children 0-18 years. There was an obvious difference in the use of nebulised bronchodilator, mucolytic and antibiotic therapies according to age ( Table 1) .

Elements of treatment burden and administration were only analysed for children 7-18 years as relatively few children under 6 required daily nebuliser use. The average number of nebulised treatments per day was 3 for those 7-15 years and 4 for those over 16. As predicted, the number of missed doses also increased with age. Of those 7-10 years, 30% reported missing more than one treatment a week. This figure increased to 50% of those 11-15 years and 90% of those over 16. Similarly, there was a gradation of time required for treatment according to age with 60% of those 7-15 years and over 90% of those over 16 spending more than 30 minutes per day on their nebulised medications. Patients (30%) reported frequently turning off their nebuliser before all the medication had been nebulised. While a significant proportion of the children were using the newer mesh nebulisers, approximately 70% were still using a compressor nebuliser pump and 55% of these were over 12 months old.

Interesting information about the posture of the children and young adults while using their nebulisers, as well as the timing of treatments relative to other therapies, was also obtained. The sitting position was used by 80%, while 70% delivered their nebulised antibiotic after completing airway clearance. Only 70% of those interviewed cleaned their nebuliser after every use with 13% cleaning less than once a week.

These results indicate that despite improvements in nebuliser technology, the time burden of treatment is significant and may play a role in the poor compliance of children with nebulised medications. As the number and type of inhaled medications become more common, it is important to achieve maximum benefit while ensuring time efficiency. At the time of prescription of inhaled medications it is important to provide education in respect to technique, timing of treatment and hygiene and this needs to be reassessed regularly as additional treatments are added to an individual patients' regimen. Background: Cystic fibrosis (CF) is characterised by dehydrated airway secretions that can be difficult to expectorate causing repeated infections, inflammation and scarring of the lungs. Physiotherapists employ a variety of airways clearance techniques (ACT) to aid sputum clearance aiming to maintain or improve lung function and hence slow progression of the disease. No single ACT has been identified in the literature to be more effective than any other or any better than no treatment at all, although Cochrane reviews of the literature comment on the paucity of long-term randomised control trials (Main et al. Cochrane Database Syst Rev 2005; (1):CD002011; Van der Schans et al. Cochrane Database Syst Rev 2000; (2) :CD001401). A previous analysis at this centre in 2007 of ACT versus FEV1 over a one year period concluded that no significant difference in lung function decline existed between any form of ACT or no treatment at all. The study did not however consider the effect of nebulised mucolytics and antibiotics.

Aim: To retrospectively evaluate the impact of various ACT on FEV1 decline over one year.

A retrospective analysis was performed on 156 adult CF patients. Patients were grouped according to ACT performed; patients with adherence less than 60% were grouped as non-compliant. Data were also collected regarding nebulised mucolytic and antibiotic usage. FEV1 decline over 1 year was the primary outcome measure and was recorded at annual review when the patient was considered to be stable. FEV1 decline versus ACT were assessed for normality and then analysed using a one way ANOVA. Post hoc analysis was done with a Scheffe test.

Results: There was a significant difference between ACTs (f=3.677, p=0.001) on FEV1 decline over one year. Regular treatment with the PEP mask improved FEV1 on average by 0.88% compared to the percussion and vibration group whose FEV1 declined on average by 3.98% per annum (p=0.031). No other treatment group differed significantly from each other. There were no differences in nebulised mucolytic and antibiotic usage between groups.

Discussion: The current analysis would suggest that PEP may be preferable to percussion in the long-term treatment of CF. These results concur with McIlwaine et al.(J Pediatr 1997; 131:570) who suggested that over one year PEP in combination with percussion and vibrations improved FEV1 more than percussion and vibrations alone in which FEV1 declined. However, in the present study the percussion and vibration group demonstrated much lower baseline FEV1 than any other treatment group. A lower baseline FEV1 may mean more frequent exacerbations and faster decline in lung function. Some treatment groups such as the autogenic drainage (AD) and Acapella groups only had small numbers which may contribute to statistical error. Additionally, adherence was reported by patients and assessed by the physiotherapist but not formally monitored. This may cause inaccuracies associated with self-reporting.

Conclusions: PEP may be better than percussion and vibrations in the maintenance of lung function in CF patients. Further studies with larger cohorts are required before any firm clinical judgements are formed. Recombinant human deoxyribonuclease (rhDNase) applied to patients with cystic fibrosis (CF), has been shown to improve lung function in shortterm trials, and there is some evidence that the number of pulmonary exacerbations may be reduced. However, its long-term effect on airway inflammation, improvement of airway clearing and hence, the effect on different characteristics of lung function has not yet been clearly assessed. In a retrospective observational study we evaluated the long-term effect of rhDNase on bacterial acquisition (onset of chronic P. aeruginosa infection; OCPAI), the effect on various lung function parameters such as functional residual capacity (FRC pleth ), lung clearance index (LCI), trapped gas (V TG ), specific airway resistance (sR eff ), forced expiratory indices (FEV 1 , FEF 50 ), gas exchange (PaO 2 , PaCO 2 ) and BMI in 170 children (85 males; 85 females) over an observation period ranging from 5 to 18 years of age. When treated, the patients received rhDNase 2.5 mg/day. Kaplan-Meier plots and linear mixed model (LMM) analysis were used to assess differences in OCPAI and drug efficacy on lung function respectively. In 72 patients the event of OCPAI occurred before initiation of rhDNase treatment (50% onset at age 4.9 years). In 75 patients rhDNase was never used (control group with 50% onset at age 7.3 years). OCPAI could be studied in 23 patients ON rhDNase, presenting a later OCPAI (50% onset at age 13.3 years; n.s.). Efficacy of rhDNase treatment was assessed by comparison of the slope 10 years before versus after initiation of treatment. The LCI (degree of ventilation inhomogeneities) was the only parameter demonstrating significant improvement (p=0.024) in patients with mild to moderate lung involvement (FEV 1 < -6 SDS; PaO 2 > 72 mmHg), but not in severely diseased patients. There was no effect on flow limitation (FEV 1 , FEF 50 ), bronchial obstruction (sR eff ), the degree of pulmonary hyperinflation (FRC pleth ), the degree of trapped gas (V TG ), nor on the BMI. Noteworthy, the beneficial effect of rhDNase was significantly more pronounced in younger patients (age < 10 years). Longterm treatment of rhDNase in patients with CF may significantly improve ventilation inhomogeneities, especially in patients with age less than 10 years. Improvement of ventilation inhomogeneities may be an expression of a better clearing of the bronchial tree explaining the delay in OCPAI in CF patients ON rhDNase. No long-term effect could be established on any of other lung function parameters (FRC pleth , V TG , sR eff , FEV 1 , FEF 50 , PaO 2 , BMI). This observational study showing only modest long-term effect of rhDNase on lung function in CF patients raises concerns about cost effectiveness. Objectives: To compare maximum muscle strength and power between children with CF and pancreatic insufficiency (PI) and healthy controls, age 5-to-19-yrs.

Methods: Anthropometry and DXA for body composition were measured and associated Z-scores generated. Maximum muscle strength (kilogram (kg)) with a handgrip dynamometer and power (watts (W)) via completion of 3 squat jumps on a force plate were assessed.

Results: Seventy-nine children with CF and PI (45 male and 34 female; FEV 1 = 97 ± 21% predicted; 90% normal to mildly reduced lung function (FEV 1 ≥ 70% predicted)) and 140 healthy controls (65 male and 74 female) did not differ in age (12 ± 4 vs. 11 ± 3 yrs; healthy controls vs. CF and PI). 26 ± 10 kg) and whole body lean mass-for-height Z (-0.03 ± 0.9 vs. -1.2 ± 0.9). Unadjusted right (20 ± 9 vs. 17 ± 8 kg, P=0.03) and left (19 ± 9 vs. 16 ± 8 kg, P=0.008) maximum handgrip strength and peak power (1589 ± 884 vs. 1279 ± 694 W, P=0.01) were significantly reduced in children with CF and PI compared to healthy controls. However, when adjusted for height, whole body lean and fat mass using multiple regression, together explaining 81, 80 and 89% of the variance in right and left maximum handgrip and peak power measures, respectively, performance was comparable between the groups.

Conclusions: Once adjusted for body size and composition, maximum handgrip and peak power measures are commensurable in children with CF and PI and healthy controls. The absence of performance decrements in body habitus adjusted muscle strength and power highlights the importance of improving growth status and body composition in children with CF.

Supported by the NIDDK (R44DK060302), Avanti Polar Lipids, Inc., Clinical Translational Research Center, (UL1RR024134), and Nutrition Center at the Children's Hospital of Philadelphia.

Elkins, M.R. 1 ; Dentice, R.L. 1 ; Bye, P.T. 1, 2 1. Respiratory Medicine, Royal Prince Alfred Hospital, Sydney, NSW, Australia; 2. School of Medicine, University of Sydney, Sydney, NSW, Australia Background: The modified shuttle test (MST) is an externally paced, incremental, maximal, field exercise test with 15 levels. In 2000, it was shown to be a reliable, repeatable and sensitive measure of exercise capaci-ty in adults with cystic fibrosis (CF). Given the progressive improvements in lung function and other measures of disease severity in adults with CF over the past decade, we aimed to determine whether the MST is still adequate for clinical monitoring of exercise capacity and for use as an outcome measure in research in adults with CF.

Methods: We selected a representative sample of 60 adults from our CF clinic. Fifty-four (90%) had performed at least one MST during the preceding 3-year period. We noted their best MST result during that period. We then randomly selected 25% of this cohort and used their demographic data to recruit age-and gender-matched adults who did not have respiratory disease. All 13 of these control participants performed the MST once.

Results: The participants with CF had a mean (SD) age of 28 (8) years and an FEV1 of 63 (20) percent of predicted. The control participants had a mean (SD) age of 31 (11) years and an FEV1 of 113 (10) percent of predicted. Eight (15%) of the participants with CF exceeded the fifteenth (maximum) level of the test. Six (46%) of the control participants exceeded the fifteenth level of the test.

Conclusion: The MST is inadequate for clinical monitoring of exercise capacity in adult CF clinics. It is also inadequate for use as an outcome measure in research involving representative populations of adults with CF, especially where control subjects without lung disease are included in the research for comparison.

Elkins, M.R. 1 ; Dentice, R.L. 1 ; Bye, P.T. 1, 2 1. Respiratory Medicine, Royal Prince Alfred Hospital, Sydney, NSW, Australia; 2. School of Medicine, University of Sydney, Sydney, NSW, Australia Background: The modified shuttle test (MST) is an externally paced, incremental, maximal, field exercise test with 15 levels. Although it is reliable, repeatable and sensitive, some adults with cystic fibrosis (CF) and many age-matched adults without lung disease exceed the fifteenth (maximum) level of the test. We aimed to determine whether a MST with 25 levels (MST-25) could maximally test these populations, and could do so with adequate psychometric properties.

Methods: We extended the MST by 10 levels, with the number of shuttles increased by one at each additional level. Twenty-one participants (10 healthy, 11 CF) performed the MST-25 on two occasions an average of 7 days apart. Pearson's r was used to assess the test-retest reliability of peak heart rate, peak SpO 2 , distance, duration, and peak energy expenditure. Bland-Altman methods were used to assess repeatability.

Results: The control participants had a mean (SD) age of 32 (12) years and an FEV1 of 112 (11) %pred. The participants with CF had a mean (SD) age of 30 (15) years and an FEV1 of 62 (22) %pred. Among the healthy participants, between-trial correlations (r) were 0.70 (p=0.02) for heart rate, 0.52 (p=0.12) for SpO 2 , 0.99 (p<0.001) for distance, 0.99 (p<0.001) for time, and 0.99 (p<0.001) for energy expenditure. Among the participants with CF, between-trial correlations (r) were 0.93 (p<0.001) for heart rate, 0.85 (p< 0.001) for SpO 2 , 0.97 (p<0.001) for distance, 0.97 (p<0.001) for time, and 0.97 (p<0.001) for energy expenditure. Mean differences (limits of agreement, coefficient of repeatability) in control participants were: 4 (-25 to 33, 29) bpm for heart rate; -1 (-4 to 3, 4) % for SpO 2 ; 64 (-25 to 153, 91) m for distance; 0.3 (-0.1 to 0.8, 0.5) min for time; and 0 (-3 to 4, 3) mL/kg/min for energy expenditure. Mean differences (limits of agreement, coefficient of repeatability) in CF participants were: -3 (-16 to 10, 13) bpm for heart rate; 1 (-8 to 9, 9) % for SpO 2 ; 31 (-160 to 222, 195) m for distance; 0.2 (-1 to 1.5, 1.3) min for time; and 1 (-4 to 6, 5) mL/kg/min for energy expenditure. Healthy and CF participants responded similarly.

Conclusion: The MST-25 has suitable reliability and repeatability for use in clinical monitoring and research.

Dentice, R.L. 1 ; Elkins, M.R. 1 ; Bye, P.T. 1, 2 1. Respiratory Medicine, Royal Prince Alfred Hospital, Sydney, NSW, Australia; 2. School of Medicine, University of Sydney, Sydney, NSW, Australia Background: A strategy of alternate side lying during nebulised medication delivery has been proposed as a means of improving the uniformity of drug deposition within the lungs. Although the ability of the side lying strategy has not yet been tested empirically, one drawback with this approach may be that it slows the rate of delivery. This might limit the clinical applicability of the alternate side lying strategy. Therefore, we sought to determine whether there was any disadvantage to delivery time with the alternate side lying strategy.

Method: Twenty-four adults (mean ± SD age 30 ± 9 y, 13F) with stable cystic fibrosis lung disease (FEV1 % pred 55 ± 34) inhaled 4 mL of normal saline via an LC Star™ nebuliser twice within 24 hours. During the first inhalation, participants were randomised to sit upright throughout the nebulisation period, or to alternate between left and right side lying at each minute during the nebulisation period. Participants used the other positioning regimen during their second inhalation. The nebuliser was stopped and weighed each minute until the residual volume of 0.5 mL was reached.

Results: The delivery time did not significantly differ between the sitting regimen (18.54 ± 3.80 min) and the alternate side lying regimen (17.96 ± 3.53 min), a difference of 0.58 min (95%CI: -1.40 to 0.24). There were no significant correlations between delivery time and any of the following parameters using either regimen: FEV1, FVC, FEV1 % pred, FVC % pred, or subject height (all R <0.59).

Conclusion: Alternate side lying during inhalation therapy does not prolong nebulisation time. We are investigating the effect of alternate side lying on the uniformity of deposition with radioaerosol scans in healthy participants and participants with cystic fibrosis. 1 1. Hospital for Sick Children, Toronto, ON, Canada; 2. Montreal Children's Hospital, Montreal, QC, Canada Rationale: Peak aerobic fitness (VO 2 peak) has been linked to outcome in cystic fibrosis (CF) patients. However, this parameter requires expensive equipment and expertise that is not readily available in all centers. Other fitness parameters such as peak anaerobic capacity, power measures such as vertical jump and strength measures such as hand grip may be simpler to deliver in the clinic. However, the relationship between these variables to established outcome measures such as FEV1 remains unclear.

Purpose: 1) Determine the relationship of measures of anaerobic capacity, muscle strength and power to aerobic capacity to evaluate their use as surrogate measures of fitness. 2) Determine the relationship of these physiological parameters to lung health (FEV1).

Methods: To date we have studied 66 patients (7-18yrs) with CF (31 female) from the CF clinic at the Hospital for Sick Children. Data were obtained during regular CF clinic visits. Testing was conducted only if patients were clinically stable, with no recent changes in their regular habitual physical activity in the last three months.

Measurements: FEV1 was measured according to standard spirometric techniques. Patients performed a maximal cycling test to exhaustion from which peak oxygen consumption (VO 2 peak; ml. kg -1 .min -1 ) was calculated. Peak anaerobic capacity (watts) for 10-(PP10) and 30-(PP30) sec trials was measured using an isokinetic cycle ergometer. Vertical jump (VJ) was measured (cm) as the difference between the highest point of the patients' jump and their standing reach. Patients were asked to squeeze a handgrip dynamometer (HG) with maximal isometric effort and hold for five seconds (kg). Group comparisons were performed using unpaired t-tests, correlations between different measures of physical fitness were performed using linear regression analysis.

Lung function was comparable in males and females whereas males scored significantly higher in all fitness measures; (PP10, PP30, VJ, HG; p=0.01). VO 2 peak was significantly correlated to PP10 (r=0.79, p=0.01), PP30 (r=0.88, p=0.01), VJ (r=0.72, p=0.01) and HG (r=0.73, p=0.01) . FEV1 was significantly correlated to VO 2 peak, PP10 and PP30. There was a weak correlation between FEV1 and HG and no significant correlation between FEV1 and VJ (Table 1) .

Conclusions: In this cross-sectional analysis, VO 2 peak and anaerobic capacity were significantly correlated to lung function whereas measures of power (VJ) and strength (HG) do not appear to be useful markers of disease in children with CF. A longitudinal analysis is currently under way to assess whether changes in anaerobic capacity are correlated to changes in lung function over time.

Supported by the Canadian Cystic Fibrosis Foundation. 2. Hôpital Laennec, Nantes, France; 3. Groupe Hospitalier Sud, Bordeaux, France; 4. Erasme Hospital, Brussels, Belgium; 5. Hôpital Européen Georges Pompidou, Paris, France; 6. Hôpital Foch, Suresnes, France; 7. University Hospital, Leuven, Belgium; 8. Hôpital Sainte Marguerite, Marseille, France Introduction: As a result of prolonged survival and better quality of life after lung transplantation (LT), many female LT recipients express a strong desire for pregnancy but only few cases of pregnancy following LT have been described.

Patients and methods: A survey was performed in the French and Belgian lung transplant centers between July 2006 and December 2008, looking for cases of pregnancy in LT recipients. We studied the characteristics of transplantation, pregnancy and maternal outcome.

Results: Twenty-three pregnancies occurred between 1991 and 2006 in 22 women who had received heart and lung (11 patients), lung (9 patients) or lung and liver transplantation (2 patients) from 1989 to 2004. The reason for LT was cystic fibrosis (14 women), pulmonary hypertension (5 women), Eisenmenger (2 women) and histiocytosis (1 woman). Pregnancy occurred within 5 months to 13 yrs (median 3.2 yrs) after LT at a median age of 29 (18−38) yrs. Two pregnancies had required assisted reproductive techniques.

There were 16 live births, 3 therapeutic abortions, 2 miscarriages, 1 in utero death at 22 weeks and 1 death of the mother in the first weeks of pregnancy. Two required caesarean sections. Six babies (37%) were premature but all 16 children were well at follow-up. Low birth weight occurred in 5 newborn babies (31%). Only 2 babies were breast fed.

Maternal follow-up after delivery ranged from 6 months to 16 years. Seven (30%) maternal deaths occurred: the first one of pulmonary infection 5 months after LT as pregnancy was just discovered; the second one of pulmonary embolism after therapeutic abortion had been performed because of obliterative bronchiolitis (OB); the third one 20 months after delivery in a patient whose pregnancy had been discouraged because of pre-existing OB; the fourth one, one year after a miscarriage that had occurred 6 months after LT; the 3 others after a survival of 7, 18 and 36 months postpartum in women whose pregnancies had occurred 4, 1 and 8 years after LT respectively and who developed OB after delivery.

OB developed after pregnancy in 5 of the 15 patients who are alive. Two of these women were successfully retransplanted (2 and 6 years after delivery). The other 13 patients are well with a mean FEV1 of 77±20% predicted.

Four children became orphans at 7, 18, 20 months and 3 yrs old. The 13 other children are 7.5±4.1 yrs old.

Conclusion: These data show that pregnancy after LT is possible but can be risky. It should not be recommended before 2 years of stable condition without OB after LT. Women and couples should be given appropriate information.

Cowperthwaite, C.; Dyce, P.; Ledson, M.J.; Walshaw, M.J.

Regional Adult CF Unit, The Liverpool Heart and Chest Hospital, Liverpool, United Kingdom Referral for lung transplantation is a major event for patients and families with cystic fibrosis (CF) that can have a profound effect on their appreciation of the CF condition and mortality. To study this further we looked at the impact of such a referral on CF patients attending our large adult unit who were subsequently accepted onto the active transplant list. We looked at how the patients had been approached when transplant referral was considered, who had approached them, whether they had been supported through the process of transplantation assessment, and their impressions of the transplant concept.

Nine patients were interviewed by one CF specialist nurse (CC) using a structured questionnaire.

All patients had first been approached by the CF specialist doctor regarding the possibility of transplant: in one case accompanied by a specialist CF nurse. In every case, the initial discussion took place in the outpatient clinic. Seventy-eight percent of patients agreed that it should be the CF specialist doctor broaching the subject, although one would have preferred the CF specialist nurse to introduce the idea and a further patient would have preferred a dual approach.

The majority of patients (89%) had an initial negative reaction when transplant was suggested: the reasons for this included shock, fear of the operation, and family experience (one patient had a CF cousin who had died post-transplant). The positive reactions experienced at this first consultation included relief that it was possible to have another treatment option, and relief that the concept of transplantation was now "out in the open."

All patients were offered the opportunity to be escorted by a CF specialist nurse to their first assessment visit to the transplant center, and 78% took this up -all found it beneficial. All patients stated they had adequate information to prepare them for this visit.

There was a mixed response regarding the patients' appreciation of the transplant team, varying from "confident in their care," "well informed," and "helpful team," to "feeling lonely," "needing more time for discussion," and "not knowing the team well enough at a crucial stage." Ideas given by patients for improved support and care included more time to discuss pre-and post-operative care, improved care of CF-related diabetes, more time to get to know the transplant team, and the need to have a post operative plan of care.

Bilateral lung transplantation for end stage pulmonary disease is a daunting and uncertain process for adults with CF. Patients prefer to be informed about the consideration for this by their CF specialist doctor, with support from CF specialist nurses. Patients may experience extreme emotions at this time, and support given at the early stages may contribute to an acceptance of their state of health and give a positive start to what can be a lengthy and complicated process.

Furthermore, the referring CF team and the receiving transplant team should liaise closely with each other, to optimise patient support and thus improve the quality of care. A coordinated systematic model should be utilised to facilitate the process which is stressful for these otherwise terminally ill patients. Denk, O.; Keller, M.; Prante, S.; Gruber, F.; Fuchs, C.; Knoch, M. Aerosol Research Institute, PARI Pharma GmbH, Munich, Germany Bronchiolitis obliterans (BO) is a major cause of death after lung transplantation, being treated today with systemic immunosuppressive drugs, e.g. cyclosporin A (CsA). Since systemic immunosuppression is of limited efficacy to prevent or treat BO and since systemic CsA has severe nephrotoxic effects, direct intrapulmonary CsA delivery may result in higher local drug exposure and better effect in association with less systemic side effects. Reports on positive therapeutic effects upon nebulisation of CsA dissolved in propyleneglycol (CsA-PG) indicated that inhaled CsA may be advantageous. Still, treatment compliance to inhaled CsA was poor due to the irritating organic solvent and long nebulisation times of up to 1 hour (FDA Briefing Document for Pulmonary Advisory Committee June 6, 2005). Thus, development was needed for a well-tolerable inhalation formulation in a low, non-toxic nominal dose but high peripheral lung dose (about 10 -28 mg/week) enabling short delivery times. A novel aqueous liposomal cyclosporin A 0.4% (L-CsA) formulation derived from an artificial lung surfactant carrier showed good cell tolerability in a Calu-3 epithelial cell culture model and was non-toxic in rats when L-CsA was given by inhalation for 6 months. Lung deposition in 12 lung transplanted patients was about 40% after inhalation of 10 mg L-CsA via a customized eFlow ® electronic nebulizer and 50% were found in the lung periphery (Behr et al. J Aerosol Med Pulm Drug Deliv 2009; 22:121) . Compared to CsA dissolved in propyleneglycol a sustained release profile could be observed for L-CsA after incubation with human lung cell homogenates. A favourable drug distribution upon inhalation was obtained in a perfused rabbit lung model with a distribution profile pattern after 6 hours as follows: 77.6% in lung homogenates, 16.4% in the bronchoalveolar lavage and 6% in the perfusate. Upon nebulisation no drug was leaking out of the unilamellar liposomes having a size range of about 50 -100 nm. Based on these positive outcomes Orphan Drug Designations were granted for inhaled L-CsA in the USA and Europe. A currently ongoing phase 2b multicenter trial will hopefully prove the efficacy and safety of inhaled L-CsA delivered via an investigational customized eFlow ® electronic nebuliser. It is our hope that detoriation in lung function caused by BO in lung transplanted patients will be reduced and their life expectancy prolonged.

Knoop, C. 1 ; Dumonceaux, M. 1 ; Ruiz, M. 2 ; Rondelet, B. 2 ; Estenne, M. 1 1. Chest Medicine, Hôpital Erasme, Brussels, Belgium; 2. Thoracic Surgery, Hôpital Erasme, Brussels, Belgium Actuarial survival 10 years after lung transplantation (LTx) is approximately 30% according to the data from the 2008 ISHLT Registry. Published data on lung function and metabolic complications in LTx recipients having survived more than 10 years are, however, scarce. We reviewed these data in our cohort of 318 LTx recipients (n=90 with cystic fibrosis (CF)) transplanted from August 1983 onwards. Thirty-seven patients survived for more than 10 years. The indications for LTx in these patients were CF (n=12), emphysema (n=8), primary pulmonary hypertension (n=5), Eisenmenger's syndrome (n=5), and other (n=7). One of 12 CF long-term survivors died 12 years after LTx (Clostridium difficile colitis, acute renal insufficiency and pulmonary emboli); the other 11 patients are currently alive. There are 8 men and 3 women. They present the following CFTR mutations: deltaF508/deltaF508 (n=4), deltaF508/3849+10kb C/T (n=3), deltaF508/G542X (n=1), deltaF508/--(n=2), unknown (n=1). Their median age is 46 (± 2.8) years. All 11 patients present exocrine pancreatic insufficiency; the median BMI is 18.4 (± 1.1). Eight patients have received a heart-lung and 3 a bilateral lung transplant. They are alive 13.9 (± 2.5) years after transplantation. At the most recent evaluation, median forced vital capacity (FVC), forced expiratory volume in one second, forced expiratory flow at 25% to 75% of FVC are 88 (± 18), 83 (± 24), 83 (± 35) % of predicted, respectively. One of the 8 CF heart-lung Tx recipients had to undergo heart re-Tx. Nine of 11 patients present treated arterial hypertension. Two patients had to undergo kidney Tx; the median serum creatinine and the median glomerular filtration rate in the remaining 9 patients are 1.5 (± 0.81) mg/% and 42 (± 31) mL/'/1.73 m 2 , respectively. Eight of 11 long-term survivors present insulin-requiring diabetes. Additionally, recurrent and/or severe Clostridium colitis has proven to be a cumbersome complication (n = 4). In conclusion, 1) 15-year survival with excellent lung function can be observed in some CF LTx recipients, 2) cardio-vascular and metabolic complications are prominent. Allergy, and Critical Care Division, University of Pittsburgh Medical Center, Pittsburgh, PA, USA; 2. Psychiatry, Psychology, Epidemiology and Biostatistics, University of Pittsburgh, Pittsburgh, PA, USA; 3. Heart, Lung, and Esophageal Surgery Institute, University of Pittsburgh Medical Center, Pittsburgh, PA, USA; 4. Cleveland Clinic Foundation, Cleveland, OH, USA Rationale: Lung transplantation (LT) is a viable but suboptimal option for patients with cystic fibrosis (CF) and end-stage lung disease. Survival in CF after lung transplantation is limited primarily by chronic rejection (BOS) and infections, which contribute to one and five year survival rates of ~80% and 55%, respectively. Induction therapy with lymphodepleting agents has emerged as a strategy to reduce rejection rates, reduce morbidity, and increase survival in solid organ recipients. In particular, induction therapy with alemtuzumab (Campath-1H) has been associated with reduced morbidity and mortality at 6 months after LT. Whether these benefits are maintained, particularly in CF patients with chronic respiratory pathogens, is unknown. We therefore reviewed survival and rejection rates in CF patients who received induction therapy with alemtuzumab at the time of LT at a single, large lung transplant center (UPMC).

Methods: We retrospectively analyzed a prospectively collected database of LT recipients with CF aged >18y transplanted at our center from 2003 (first year of alemtuzumab use) to 2009 (n=62, age range 19 to 57y). Survival rates were analyzed for the entire cohort, and the three year rejection-free survival rates determined for the first 43.

Results: All patients underwent T cell depletion with alemtuzumab followed by immunosuppression with tacrolimus and low dose (5 mg/day) prednisone +/-mycophenolate mofefil. Valganciclovir and voriconazole were used routinely for prophylaxis against CMV and fungal infections. Inhaled antibiotics were used routinely for the first few months depending on pre-transplant respiratory cultures. Surveillance bronchoscopy was performed at 2 weeks, and every 2-3 months for the first two years after transplant, and as clinically indicated. The median lung allocation score was 39.6 (range 32 to 90). Two patients underwent synchronous double lung and liver transplants, and 3 patients had Child's A liver cirrhosis at time of lung transplant. Nine of the 62 patients (14.5%) were on invasive mechanical ventilator support at time of transplant, and 10 had Burkholderia species (5 cenocepacia (Bcc)) in pre-transplant cultures. Excluding patients with Bcc, one, three and 5 year actuarial graft/patient survival were 98.1%/98.1%, 74.2%/80.3% and 65%/71.4%, respectively. One year survival for patients with Bcc was 60%. Three patients underwent re-transplantation at 16, 23, and 35 months after initial transplant. Freedom from grade A2 or higher Acute Cellular Rejection at three years was 39.7%, and actuarial BOS-free survival at 3 years was 71.5%. BOS and infection with Bcc or fungi were the major causes of graft failure or death.

Conclusions: Alemtuzumab induction is associated with improved short-and intermediate-term outcomes in LT for a high risk CF population. Survival and freedom from acute and chronic rejection appear to be improved relative to the international lung transplant registry and published series.

Blackwell, L.S.; Romero, S.L.; Barker, D.H.; Grimley, M.E.; Quittner, A.L. Psychology, University of Miami, Coral Gables, FL, USA Objective: Lung transplantation is a difficult decision for patients, families, and health care teams. Providing empirical information about healthrelated quality of life (HRQOL) may provide useful information in this process. Unfortunately, few studies have examined post-transplant quality of life in cystic fibrosis (CF), and to our knowledge, none have examined it prospectively. The current prospective study assessed children's and adolescents' self-reported emotion, social, and role functioning, using a standardized HRQOL instrument, the Cystic Fibrosis Questionnaire-Revised (CFQ-R).

Methods: Seventy-five pediatric patients (M=age 13.89 years; 64% female) and their parents were recruited from centers belonging to the International Pediatric Lung Transplant Collaborative (IPLTC). All patients were listed for transplantation. Participants completed the CFQ-R every six months prior to transplantation and approximately every three months following transplantation. Because each child had a different number of assessments pre-and post-transplantation, mixed modeling was used to estimate differences on each CFQ-R domain (Emotional, Social, and Role Functioning) from pre-to post-transplantation.

Results: Preliminary analyses indicated that children and adolescents reported significant improvements on all domains. Emotional Functioning, as reported by patients and parents, was low prior to transplant (M=67.05, M=66.12, respectively), with significant improvements reported after transplant by both patients (M=79.01) and parents (M=81.01). Similar improvements were found pre-to post-transplantation for patients on the Social (pre M=61.99; post M=75.23) and Role Functioning scales (pre M=59.86; post M=77.75).

Discussion: Prior to transplantation, CF patients' disease severity had a significant impact on their emotional functioning, interaction with peers and performance of social roles (e.g., school, work). After transplantation, patients reported significant improvements in all of these domains. These preliminary data suggest that transplantation improves patients' psychosocial functioning. Future analyses should use a Q-Twist analysis to examine HRQOL and patient survival in patient pre-and post-transplantation.

Funding provided by the Cystic Fibrosis Foundation. Objective: This study evaluated both children's and parents' perceptions of Treatment Burden before and after lung transplantation, using a standardized health-related quality of life (HRQOL) instrument. Relationships between treatment burden and disease severity prior to and following transplantation were also examined.

Methods: Participants included 75 children and adolescents with cystic fibrosis (CF) (Mean age = 13.89 years; 64% female) and their parents, recruited from centers belonging to the International Pediatric Lung Transplant Collaborative (IPLTC). Pulmonary function (FEV1% predicted; Mean = 33.48%), was used as a marker of disease severity. Participants reported their perceptions of Treatment Burden (0 to100, higher scores indicate less burden) on the Cystic Fibrosis Questionnaire-Revised (CFQ-R) during clinic visits before transplant and after transplant.

Results: Preliminary analyses indicated that prior to transplantation, Treatment Burden was considerable as reported by both patients (Mean= 51.25) and parents (Mean = 43.86). Significant improvements in Treatment Burden were reported by both patients (Mean= 78.50) and parents (Mean = 83.07). Prior to transplant, Treatment Burden was not related to FEV1% (Children, r = -.09, parents r = .07). Thirty-three patients received a trans-plant. After transplantation, no significant associations were found between FEV1% predicted and Treatment Burden for patients (r = .11). However, for parents, a significant relationship was found (r = .47).

Conclusions: Prior to transplantation, both patients and parents reported significant levels of treatment burden. For those patients who received a transplant, Treatment Burden improved substantially. Few associations between pulmonary function and Treatment Burden were found. However, after transplantation, parent caregivers' scores on Treatment Burden were significantly correlated with pulmonary function. This may reflect some type of "vigilance" on the part of caregivers, who may be closely monitoring their children's/adolescents' lung function in relation to daily treatments. Future analyses will use longitudinal modeling to examine changes in HRQOL from pre to post-transplantation.

Funding was provided by the Cystic Fibrosis Foundation.

Background: Although newborn screening for CF (NBSCF) is included in many routine newborn screening (NBS) programs, there is no ideal newborn screening test.

Aim of study: To assess the test performance of 2 new strategies for NBSCF.

Methods: In 2008 NBSCF was added to routine NBS in a part of the Netherlands. Screening for CF was performed by 2 strategies in each heel prick sample sent for analysis in 2 of the 5 Dutch screening labs. The first strategy used immunoreactive trypsin (IRT) and pancreatitis-associated protein (PAP). The test was considered positive with concentrations of IRT≥50 and PAP≥1.8 or IRT≥100 and PAP≥1.0 (all µg/l). For the second method, samples with IRT ≥50 were analyzed for 36 CF-mutations. An extended gene-analysis (EGA) was performed in each sample with 1 CF-mutation. Tests were positive only when 2 CF-mutations were identified.

Results: With the IRT-PAP strategy 119 of the 72874 screened neonates had a positive test, among those 10 CF patients were identified, 1 CF-patient with meconium-ileus was missed. With the IRT-DNA-EGA strategy 20 positive tests identified 11 CF-patients and 9 compound heterozygotes (7 with R117H-7T as 2nd mutation). IRT-DNA-EGA identified 89 carriers. Patients with meconium ileus were excluded for the analysis of the test performance. Sensitivity was 100% for both strategies. Specificity and PPV were 99.85 and 8.4% for the IRT-PAP, respectively 99.99 and 53% for the IRT-DNA-EGA. Based on the results of both strategies we calculated a sensitivity of 100%, specificity of 99.997% and PPV of 83% for a combined strategy IRT-PAP-EGA.

Conclusion: Compared to current NBSCF-strategies, an improved test performance of the IRT-DNA-EGA strategy was found, but not of the IRT-PAP strategy. With the IRT-PAP strategy no carriers were identified and no infants with an equivocal CF-diagnosis. IRT-PAP-EGA may be an even better strategy, combining the best test qualities of both strategies.

Castellani, C. 1 ; Tamanini, A. 2 ; Capodaglio, A. 1 ; Rizzotti, P. 2 ; Assael, B.M. 1 1. Ospedale Civile Maggiore, Cystic Fibrosis Center, Verona, Verona, Italy; 2. Ospedale Civile Maggiore, Verona, Italy In Veneto and Trentino Alto-Adige, two regions located in the northeastern Italy, CF Neonatal Screening has been running for more than 30 years. The protocol currently in use started in 2003, and adopts a three tier system, whose progressive steps are immunoreactive trypsinogen (IRT) measurement, mutation analysis with complementary meconium lactase determination, and sweat test. The aim of this study was assessing the performance of this strategy.

An IRT level equal to the 99.5th centile is used as the cut-off point. At or above this value, mutation analysis and meconium lactase determination are performed. From January 1993 to February 1995 the mutation panel included F508del, R1162X and N1303K, estimated to cover 61% of the affected alleles in the area; from March 1995 10 other mutations were included, thus reaching an 85% mutation detection rate. Later more mutations were progressively added to the panel, which now includes 34 mutations. Diagnosis is established, and later confirmed by a sweat test, in homozygous or compound heterozygous neonates. A sweat test is performed also if either one mutation tests positive or if the meconium lactase exceeds a given threshold. As a fail-safe strategy, when IRT at birth is above the 99.9th centile, even if lactase and mutation analysis test negative, a second blood sample is requested: a persistently raised IRT calls for a sweat test. Only newborns with an undisputable CF diagnosis, defined by sweat chloride above 60 mmol/L, were included.

Between January 1993 and December 2008, 838,510 newborns were screened for CF, with an average per year of 52,406. IRT values above the threshold were found in 4,451 neonates (0.53%), of whom 295 had high meconium lactase activity, 352 at least one mutation, 145 both raised meconium lactase and one or two mutations, and 86 persistently raised IRT. Altogether, 878 newborns were sweat tested. CF was diagnosed in 202 newborns. By eliminating meconium lactase testing, we would have missed 3 diagnoses (elevated IRT, -/-mutation analysis, negative lactase). Nine children with CF were missed by the neonatal screening procedure, and were later diagnosed by clinical symptoms (mean age at diagnosis 49.5 months, minimum 2 maximum 139). Of the false-negatives, 8 had IRT values below the cut-off, one had raised IRT but lactase was in range and the genetic analysis wrongly resulted negative. The overall incidence of CF was 1/4,151, which raised to 1/3,973 including false-negatives. Each case was classified as: true positive (TP); false positive (FP); true negative (TN); or false negative (FN). The sensitivity of the program, defined as TP / (TP + FN), was 0.957. The specificity of the program, defined as TN / (TN+FP), was 0.999. The positive predictive value of the test, defined as TP /(TP + FP), was 0.23. The ratio of sweat tests requested to true CF patients was 4.34:1. In Mediterranean populations CFTR mutations are quite diversified, and even specifically designed mutation panels may obtain unsatisfactory detection rates. The protocol we are using allows good sensitivity and specificity. Methods: Infants included in this analysis were identified by CA CF NBS which began in July 2007 and has 4 steps: 1) measuring immunoreac-tive trypsinogen (IRT) levels in all newborn blood spot specimens, 2) specially designed CA (29-40) CFTR mutation panel testing of specimens with high IRT values (≥62 ng/mL, top 1.5%), 3) CFTR full gene sequence analysis utilizing scanning-sequencing technology of specimens found to have only 1 mutation in Step 2, and 4) referring infants with 2 mutations found during Step 2 ("Step 2 positives") or 1 mutation found during Step 2 and 1+ mutation/variant found during Step 3 ("Step 3 positives") to CF Care Centers (CFCs) for SC testing and long term follow up unless mutations are in cis. CFCs attempted to repeat [SC] on all infants with initial [SC] ≤59 mEq/L. We analyzed [SC] and diagnoses for all Step 3 positives and stratified the analysis by initial [SC] ("abnormal" (≥60 mEq/L), "intermediate" (30-59 mEq/L), and "normal" (≤29 mEq/L)).

Results: In the first year, 42

Step 2 and 168

Step 3 screen positives and 289 carriers were identified. All but one of the Step 2 positives had abnormal initial [SC] and were diagnosed with CF. None of the screen negative carriers have been diagnosed with CF, so far. Among Step 3 positives, initial [SC] was abnormal in 17 (10%), intermediate in 21 (12%), normal in 117 (70%), not done in 8 (5%) due to repeated insufficient quantity, infant death, or parent refusal, and 5 (3%) were excluded after mutations were found in cis. All infants with abnormal initial [SC] Methods: Cox proportional hazards regression models were used to compute hazard ratios (HR) and 95% confidence intervals (CIs) testing the effect of baseline (fixed) covariates (e.g., genotype, gender, age at CF diagnosis) and time-varying covariates (e.g., tobacco smoke exposure, infection with other CF pathogens) on the risk of early Pa acquisition. The censored failure time was age at first isolation of Pa while on study, and participants remaining Pa negative throughout the study were right censored at the time of their latest follow-up. As of December 2008, the cohort consisted of 1704 patients from 59 sites who met the following inclusion criteria: 1) diagnosis of CF; 2) age ≤12 years; 3) no prior isolation of Pa from a respiratory culture ("Pa-never," N=1120), or, if prior isolation of Pa, at least a two-year history of Pa negative cultures ("Pa-past," N=584).

Results: As of December 2008, 845 (49.5%) of the cohort had converted to Pa+, including 506 (45.1%) of the Pa-never subjects and 339 (58.0%) of the Pa-past subjects. Pa conversion rates were 0.22 per person-year among the Pa-never subjects and 0.30 among the Pa-past subjects (p<0.0001). In separate univariate analyses, among Pa-never subjects, the following characteristics were significant risk factors for earlier Pa acquisition: high-risk CF genotype (both CFTR mutations in functional class I, II or III; N in model = 1059, HR 3.35, 95% CI 2.09, 5.38 ) and persistent iso-lation of S. aureus from respiratory cultures (defined as ≥3 cultures + for S. aureus during the observation period, N = 239, HR 1.92, 95% CI 1.48, 2.50) . The following characteristics were found to significantly decrease the risk of early Pa acquisition: diagnosis by prenatal or newborn screening (N = 964, HR 0.72, 95% CI 0.57, 0.91) , and, among children ≤3 years of age, breastfeeding in the year prior to enrollment (N = 497, HR 0.75, 95% CI 0.56, 0.97). Among Pa-past subjects, the only significant risk factor was persistent isolation of S. aureus from respiratory cultures (N = 402, HR 2.09, 95% CI 1.52 2.87). Environmental exposures including hot tubs, swimming pools, tobacco smoke, daycare attendance, palivizumab RSV immunization and social events with other persons with CF as assessed by parent report on an annual questionnaire did not appear to be risk factors for age at Pa acquisition (with 80% power to detect a minimum HR of 1.3 at the .05 significance level).

Conclusions: In these univariate analyses in a large U.S. cohort, high risk CFTR genotype and persistent respiratory infection with S. aureus were risk factors for acquisition of Pa. In contrast, identification through prenatal or newborn screening and breastfeeding in the year prior to enrollment appeared to be protective.

Flanary, J.C. 1 ; Gondor, M. 1 ; Lehmann, S. 1 ; Benford, S. 2 ; Hickstein, R. 2 ; Gould, J. 2 1. The CF Center at All Children's Hospital, Pediatric Pulmonary Associates, P.A., All Children's Hospital, St. Petersburg, FL, USA; 2. Department of Pathology and Laboratory Medicine, All Children's Hospital, St. Petersburg, FL, USA Background: Cystic fibrosis (CF) is a genetic disorder that primarily affects the lungs and digestive system resulting in thick mucus production that blocks ducts in the pancreas preventing normal transport of trypsinogen. Immunoreactive trypsinogen (IRT) levels are analyzed by the Florida Newborn Screening Program to identify infants at an increased risk for CF.

Each day, the top 4% of elevated IRT values are screened for 39 common, known CF mutations. Of these elevated values, those with at least one mutation for CF, or with an IRT > or = to 170 are referred to a CF Center for diagnostic sweat testing.

Method: Sweat was collected using the Wescor Macroduct® System. If samples were collected from two sites, the sweat analysis was performed in duplicate and the results averaged. A negative sweat chloride is <30 mnmol/L; an intermediate level between 30 and 59 mmol/L; a positive sweat chloride > or = to 60 mmol/L.

Results: Of 102 patients evaluated, 12 were confirmed as having CF by sweat chloride results, at least one identified mutation, and clinical symptoms consistent with CF. These 12 patients had a mean IRT value of 164.79 with one standard deviation of 73.46. All 12 patients had IRT values >93 ng/mL. There were 7 patients with IRT values >93 ng/mL, but with negative sweat chloride results. One patient with no identified mutations had an elevated IRT and a sweat chloride at the low end of intermediate. Eighty-one patients had IRT values <93 ng/mL and negative sweat chloride results. There were 90 out of 102 patients referred to our center with negative sweat chloride results.

Conclusion: Statistical analysis showed that an IRT value greater than 84 ng/mL (99.9% confidence limit) would indicate that a newborn is at risk for CF. Using 84 ng/mL, the false-positive rate in this analysis would be 14.7% (15 false-positive IRT/102 patients) or 85.3% specificity. The falsepositive rate without a cut-off would be 88.2% (90 false-positive IRT/102 patients) or 11.8% specificity. The 84 ng/mL cut-off would appear from this data set to offer improved specificity without sacrificing sensitivity. The N value in this analysis is low and data will continue to be collected and analyzed.

Please see CF positive patients shown in dark circles on the graph below.

Stephenson, A.L. 1 ; Tullis, E. 1, 4 ; Austin, P. 2 ; Ray, J. 2 ; Corey, M. 3 ; Hux, J. 2 1. Adult CF Program, St. Michael's Hospital, Toronto, ON, Canada; 2. The Institute for Clinical Evaluative Sciences, Toronto, ON, Canada; 3. SickKids Hospital, Toronto, ON, Canada; 4. Li Ka Shing Knowledge Institute, Toronto, ON, Canada Background: Low socioeconomic status (SES) has been consistently linked to worse health outcomes in the general population. Within the cystic fibrosis (CF) population in Canada, where health care is universally available, no studies have examined how SES and rural residence affect CFrelated outcomes.

Purpose: To determine the effect of SES and geography on hospitalization rates in a cohort of individuals with CF.

Methods: We completed a retrospective cohort study between 1993 and 2002 of all CF patients aged ≥ 10 y and older living in the province of Ontario. Individuals within the Canadian CF Foundation Patient Data Registry (CPDR) were probabilistically linked to health care databases that detail inpatient, outpatient and homecare services. Using ICD-9 diagnosis codes, we estimated the number of hospitalizations per year for pulmonary exacerbations for males and females. Poisson regression models using generalized estimating equations methods included predictors of hospitalization, such as age, % predicted FEV 1 for children and adults, nutritional status (body mass index [BMI] in adults, CDC BMI z score for children), pancreatic status, presence of CF-related diabetes (CFRD) and microbiology. SES was based on neighbourhood income quintiles using postal code information from Census and Statistics Canada data. Rural community size was defined at a population < 10,000. The distance between place of residence and CF centre was calculated as the shortest distance (in km) using longitude and latitude measurements.

Results: A total of 1043 persons (45% female) had 3223 respiratoryrelated hospitalizations over the 10-year study period equivalent to an incidence rate of 47 hospitalizations per 100 patient-years. Among those aged 10-19 y, univariate analysis showed that SES (p = 0.94), distance to CF centre (p = 0.42) and community size (p = 0.39) were not significantly associated with hospitalization. For those aged > 19 y, community size (p = 0.87) and SES (p = 0.07) were not significantly associated with hospitalization; however distance to a CF centre was significantly associated with hospitalization (p = 0.03) in the univariate analysis. After adjustment for other clinical predictors, distance to CF centre was no longer significant (p = 0.49). Female gender, low lung function, poor nutrition, CFRD, pancreatic insufficiency and B. cepacia were all significant predictors of hospitalization in multivariate analysis in both age groups.

Conclusions: SES and geographical proximity to a CF care centre were not significant predictors of hospitalization in CF children and adults. In a universal health care system, and within a network of specialty CF clinics, care may be the same regardless of income or place of residence. However, females with CF have higher hospitalization rates, even after accounting for markers of disease severity. This may reflect gender differences in access to medical care or outpatient management, variations in symptomatology, or hormonal differences. Further study is needed to elucidate the reasons for this gender disparity.

Newborn screening (NBS) for cystic fibrosis (CF) was initiated in April 2008 in Ontario, Canada. Approximately 37% of CF NBS infants were previously reported to be pancreatic sufficient (PS-CF) at diagnosis. Our aim was to investigate the pancreatic status of infants with a confirmed diagnosis of CF (sweat chloride > 60mmol/L and/or 2 CF-causing mutations) and those with an uncertain diagnosis of CF (sweat chloride 30-60mmol/L; U-CF) who were assessed in the CF clinic at The Hospital for Sick Children in the first few months of life using stool elastase-1 (Monoclonal-ELISA. Schebo Biotech AG) and adjunctive tests (serum Vitamins A and E and serum albumin). Of 14 infants with a confirmed diagnosis of CF through NBS, 10 had stool samples tested for elastase at mean age 42 ± 36 days. Eight of the 10 infants (80%) were pancreatic insufficient (stool elastase <100 µg/g (1); PI-CF). All infants with CF had lower stool elastase compared with all 5 patients with U-CF. There were significant differences in mean stool elastase between infants with U-CF, PS-CF and PI-CF (Table) . Four PI-CF infants had low serum Vit A levels (<0.6 µmol/L), while Vitamin E was low in one PI-CF infant (< 5 µmol/L) and in the normal range for the remaining 9 infants. Serum albumin values tended to be lower in the PI-CF group at 26.9 ± 4.6 vs. 34 ± 4.2 (p = 0.08), and except for 3 patients in the PI group, all were in the normal range. None of the infants with U-CF had low serum Vit A, Vit E, or albumin levels. The degree of pancreatic dysfunction between U-CF, PS-CF and PI-CF was mirrored by the differences in sweat chloride levels (Table) .

In conclusion, there is a spectrum of pancreatic exocrine dysfunction among NBS infants with CF. NBS infants with PS-CF do not have normal pancreatic function. Instead, they have an intermediate degree of pancreatic exocrine function, in between those with PI-CF and U-CF. These findings support the role of monitoring PS-CF infants for possible progression to PI. Testing for pancreatic elastase in a timely manner helps to define pancreatic status, and guide treatment for PI. Serum Vitamin E, A or albumin may only be useful if low values are observed. 1. Beharry S et al. J Pediatr 2002:141; 84-90 . CF Center, Karolinska Institutet, Stockholm, Sweden; 3. Respiratory Medicine, Heart and Lung, Lund, Sweden; 4. Uppsala CF Center, Women's and Children's Health, Uppsala, Sweden; 5. Respiratory Medicine and Allergology, Sahlgrenska University Hospital, Gothenburg, Sweden; 6. Cystic Fibrosis Foundation, Bethesda, MD, USA; 7. Dartmouth Hitchcock Medical Center, Lebanon, NH, USA Background: Both Sweden and the USA are affluent countries with high standards of living. In contrast to the USA, Sweden has a taxpayer supported health care system with minimal "out-of-pocket" costs. The CF patients are seen every 4-6 weeks. This might influence the approach towards different treatment choices in CF. As part of a pilot study evaluating the potential use of PortCF as a European patient registry, the 4 Swedish centres filled in one full data set during 2007. The aim was to use the information to describe differences in treatment and outcome.

Method: Information about social data, lung function, anthropomorphic data and chronic treatment was available. Due to the few data sets compared to the USA entries no comparison could be made regarding number of pulmonary exacerbations and use of IV antibiotics. A few comparisons could be done regarding microbiology data.

Results: Five hundred nineteen (54.9% M, mean age 21.5±13.8, median 18.8 yrs) Swedish patients and 23 842 US patients (51.8% M, mean age 18.5±12.5, median 16.6 yrs) were included. Age at diagnosis was similar (60% at age 12 months and 10 months) while in Sweden fewer patients were treated with pancreatic enzymes (80 vs 90% respectively). Adult patients worked full-time to a similar degree (38% and 37%) while part-time work was more common in Sweden (26% vs 11%). The median BMI percentiles for children ≤19 yrs were 49 and 47 while BMI in adults were 21.9 and 21.7 in Sweden and the USA, respectively. The median FEV1.0 was 96.5 and 92.5% (children), and 75.2 and 63.9% (adults) in Sweden and the USA, respectively. The higher percentage of patients with pancreatic sufficiency in Sweden was not accounted for in the analyses.

Treatment data: The dosage of pancreatic enzymes was the same in children (1672 vs 1690 units of lipase/kg/meal) while it was higher in US adults (1002 vs 1458 lipase/kg/meal). The use of acid blockers was higher in the US (children 15% vs 61%, adults 21% vs 49% respectively).

The use of rhDNAse (≥6 yrs), inhaled tobramycin (TOBI) (in patients with chronic PA) and inhaled steroids (in patients with no asthma) was lower in Sweden compared to the USA (15.5 vs 74%, 20.8 vs 66.5%, and 18.3 vs 49.1% respectively) while using exercise as primary or secondary airway clearance technique was higher in Sweden (68.2 vs 24.5%).

Sweden had less multiresistant P. aeruginosa, less S. aureus in the younger age groups and less MRSA.

Conclusion: Large differences in treatment approaches were noted despite similar outcomes in nutritional status and lung function, especially in the pediatric population. The study was not designed to answer why there are differences, but raises questions of how optimal CF care should be performed and how and if the organisation of the health care influences treatment decisions and long term outcomes. Quittner, A.L. 1 ; Rasouliyan, L. 2 ; McMullen, A. 3 ; Wagener, J. 4 ; Woo, M.S. 5 ; McColley, S. 6 ; Sawicki, G.S. 7 1. University of Miami, Miami, FL, USA; 2. ICON Clinical Research, San Francisco, CA, USA; 3. University of Rochester School of Medicine, Rochester, NY, USA; 4. University of Colorado Denver School of Medicine, Aurora, CO, USA; 5. University of Southern California, Los Angeles, CA, USA; 6. Northwestern University Feinberg School of Medicine, Chicago, IL, USA; 7. Harvard Medical School, Boston, MA, USA Rationale: Disease-specific, health-related quality-of-life (HRQOL) measures provide unique information in cystic fibrosis (CF). We examined whether changes in respiratory and nutritional indices predicted changes in the Cystic Fibrosis Questionnaire-Revised (CFQ-R) scores over 1 year.

Methods: Participants enrolled in the Epidemiologic Study of Cystic Fibrosis (ESCF) who completed age-specific CFQ-R assessments (Child, Parent, Adolescent, and Adult) on two occasions separated by 9 to 15 months were included. Clinical variables, including cough, crackles, wheeze, IV antibiotic treatment, and nutritional indices were used to construct multivariable linear models for each CFQ-R domain in children (6-13 years old), parents (of children), adolescents (14-17 years old), and adults (≥18 years old).

Results: 1947 pairs of CFQ-R assessments were analyzed: 337 Child (mean age 8.9 years), 581 Parent (mean age of child 8.8 years), 398 Adolescent (mean age 15.3 years), and 631 Adult (mean age 26.9 years). The average time between assessments was 12.1 ± 1.5 months. Changes in respiratory signs and symptoms were associated with significant changes in Physical, Vitality, and Respiratory domains for adults, adolescents, and parents (beta coefficients ranging from -3.04 to -9.90, P values <.05). New symptoms such as cough, wheezing, or IV antibiotic treatment were associated with worsening CFQ-R Respiratory Symptom scores; however, different respiratory symptom patterns emerged across age groups. For children, increased cough was predominant; for adults, changes in cough, crackles, wheeze, and IV antibiotic treatment clustered to predict changes in HRQOL scores. Changes in weight-for-age z-scores were associated with significant changes on the Body Image, Eating Problems, and Weight domains across the groups (beta coefficients ranging from 6.38 to 26.71, P values <.01). Improvements in weight were associated with better perceptions of body image and fewer eating problems.

Conclusion: In this CF population, changes in pulmonary signs and symptoms, and nutritional status were predictive of substantial changes in HRQOL domains over 1 year. CFQ-R scores add important information to traditional health outcomes and may serve as an additional tool for clinicians making treatment decisions.

Acknowledgments: This study was supported by Genentech, Inc., South San Francisco, CA. The authors gratefully acknowledge the patients, parents, investigators, and coordinators of the Epidemiologic Study of Cystic Fibrosis (ESCF).

Lang, C.W. 1 ; McColley, S. 2 ; Ross, L.F. 1 1. University of Chicago, Chicago, IL, USA; 2. Children's Memorial Hospital, Chicago, IL, USA Objectives: Newborn screening (NBS) for CF began in March 2008 in Illinois. Illinois defines a positive NBS as 1) an elevated IRT with at least one DNA mutation using a 44-mutation panel; and 2) ultra-high IRT without a known mutation. We studied maternal knowledge and attitudes to a positive NBS and a normal sweat test.

Methods: Sweat tests were conducted at 2 accredited CF centers which offer concurrent genetic counseling. After receiving sweat test results, participants were invited to be contacted for this study. Those who consented participated in a phone survey that included 20 knowledge questions about the genetics and symptoms of CF, experience with CF and CF testing, attitude about disclosure of carrier status, and demographics.

We interviewed 30 women ages 18-40 years (mean 29±6 years). Most were married (20, 67%), Caucasian (16, 53%), with private health insurance (18, 60%). Almost half (14, 47%) were college graduates. Sixteen (53%) of sweat tests took place <3 weeks after birth. Two-thirds (19, 63%) had heard about CF before the NBS but only 9 (30%) discussed NBS with their obstetrician (OB). Only eight (27%) discussed CF carrier testing with their OB. Twelve (40%) knew that the gene for CF was present in their or their partner's family. Approximately 75% were initially notified of the NBS result by a pediatrician, and were told about CF and the meaning of a positive NBS for CF at this time. However, 17 (57%) were not given a description of the sweat test. Reported anxiety was a 3.40 (out of 4) at initial notification. Three-quarters spoke to a genetic counselor when scheduling the sweat test. Three did not schedule the sweat test themselves, but of the remaining 27, over 85% received information about CF, the NBS, and the sweat test when scheduling. Their reported anxiety decreased to 2.72 after phone counseling. All but one correctly understood the interpretation of their child's sweat test; however 10 (33%) did not understand if a mutation was found and 9 (30%) did not correctly understand their child's carrier status. Reported anxiety after receiving sweat test results was 1.25. Despite this, 9 (30%) reported thinking about the results once a week or more. The mean score for knowledge of CF was 78%: 100% understood that CF cannot be transmitted by physical contact but less than 1/2 understood that there are no health risks associated with being a carrier. Among 24 mothers who identified their children as carriers or potential carriers, ~90% think that first degree relatives should know that their child is a carrier; but <50% think that other relatives should know, and 3 (13%) do not want other relatives to find out. All plan to tell the child that he/she is a carrier; the majority (79%) plan to wait until the child is >13 years.

Conclusions: Most women were not offered prenatal CF carrier testing. Most OBs also failed to discuss NBS. Initial notification of an abnormal NBS for CF creates a high level of anxiety, but the process of speaking to a genetic counselor when scheduling a sweat test provides reassurance before the sweat test is performed. Knowledge of CF after genetic counseling is good, but confusion about residual risk persists.

Funding: Illinois Department of Public Health. In April 2008, the province of Ontario, Canada, expanded newborn screening to include testing for cystic fibrosis (CF). Screening involves the measurement of immunoreactive trypsinogen (IRT), and mutation analysis with a panel of 39 common mutations as second tier testing. Screen positive infants are stratified based on the presence or absence of mutations +/-significantly elevated IRT. During the period of April 2008 -April 2009, 229 screen positive infants were identified in our center. Of these infants, 12 have been diagnosed with CF.

To address the technical complexity and unique psychosocial issues raised by screen positive results for CF, the NBS Centre at our Hospital has partnered closely with the CF clinical team to address the needs of these families. The NBS team has expertise in the disclosure of screen positive results, and provides appropriate education, support, and resources to families throughout the follow-up process. The clinical CF team brings expertise in medical management, education, treatment, and psychosocial support. By merging our two interprofessional teams, we are able to provide a model of care that promotes seamless transition and continuity of care for neonates confirmed to have CF.

With direction from our medical lead, the team developed a clinical pathway that guides the retrieval and follow-up process. When the NBS team is alerted to a screen positive case, the genetic counsellor locates the family and/or health care provider to disclose the results and plan for followup by sweat chloride testing. Sweat test result disclosure, counselling, education, and the option of parental carrier testing are facilitated, same day, by the CF team nurse practitioner and/or NBS genetic counsellor. This approach provides families with access to expertise in clinical aspects of the disease, simultaneously with counselling regarding broader genetic implications for the infant and family. For families, this model of follow-up maximizes the expertise and unique perspectives that an interprofessional team approach provides, serving to promote positive patient outcomes.

Walker, W.T. 1 ; Connett, G. 1 ; Kay, H. 2 ; Legg, J. 1 1. Regional Paediatric Cystic Fibrosis Unit, Southampton University Hospitals Trust, Southampton, United Kingdom; 2. South & West Cystic Fibrosis Database, Royal United Hospital, Bath, United Kingdom Introduction: Allergic bronchopulmonary aspergillosis (ABPA) is a lung hypersensitivity disorder mediated by an allergic late-phase inflammatory response to certain antigens of Aspergillus fumigatus. It is recognised as an important complication of cystic fibrosis (CF), typically presenting with wheezing, transient pulmonary infiltrates and reduced lung function. Increasingly aggressive antibiotic treatment regimes have raised concerns that such approaches might select out increased fungal growth and allergic sensitization.

The South and West Cystic fibrosis Database was established in 1995 and is managed by a well-resourced team that collects and verifies detailed clinical information on all CF adults and children within our region. The database currently collects data on over 900 patients with CF cared for in 23 hospitals. The most recent epidemiological reports on ABPA in the literature are based on data from over ten years ago. We report on data from the South and West database for the period 2002-2007.

Results: The annual incidence of ABPA (new diagnoses of disease occurring within the CF population) has remained comparatively constant during the years studied varying between 2.25% and 2.94% per annum. The annual prevalence of ABPA (total number of individuals with ongoing disease within the CF population) has decreased slightly year-on-year since its peak in 2003, from 5.96% to 4.29% in 2007, see Table 1 . There was a predominance of males diagnosed during the study period (58.9%) although subgroup analysis demonstrated a significantly higher proportion of females than males in the 5-9 year age group. ABPA was associated with a significantly higher rate of isolation from respiratory specimens of both Pseudomonas aeruginosa (p<0.05) and Aspergillus fumigatus (p<0.01) than those without ABPA. A considerable proportion (24/134; 17.9%) of those diagnosed during the study period went on to have recurrent episodes of ABPA necessitating further treatment.

Discussion: These data represent robust current epidemiological trends on the incidence and prevalence of ABPA in CF patients in our region. Contrary to our clinical impression, our findings suggest that there has been little change in the prevalence of APBA in the CF population over recent years. The observations of a recent trend towards a reducing prevalence and of a female preponderance in early life will require ongoing data collection to confirm. Current antibiotic strategies do not appear to be associated with increased occurrence of ABPA within our regional CF services. 3 ; Cullinan, P. 3 ; Larsen, A. 2 1. Department of Respiratory Medicine, Royal Brompton Hospital, London, United Kingdom; 2. Cystic Fibrosis Trust, Bromley, United Kingdom; 3. Department of Occupational and Environmental Medicine, National Heart and Lung Institute, Imperial College, London, United Kingdom Background: The UK cystic fibrosis (CF) Registry adopted the web based, user friendly, Port CF software 2007 onwards, in order to capture data from all CF Centres in the UK. The aim was to facilitate regional, national and international audit of key outcome measures in CF.

Methods: We identified 4408 patients with complete clinical data in 2007 from the UK CF Registry. Details of the date and method of diagnosis were collected at registration and data collected at a clinical encounter in 2007 included lung function testing, BMI and microbiological cultures. An annual review questionnaire was also completed for a subset of patients which enquired about chronic infections and current employment status. Median predicted survival was estimated among patients registered in 2007 regardless of completeness of clinical data.

Results: Of patients with complete clinical data, the median age was 18 years with an almost equal distribution among sexes (males: 53.9%); 56.7% of patients were aged 16yrs or over. Patients were diagnosed in early life (median=5 months); of the 31 patients born in 2007, 21 were identified through neonatal screening. Genotyping was performed in 92.6% of patients; most were either homozygous (54.4%) or heterozygous (37.9%) ∆F508. Among patients 16 years of age and older who completed an employment status questionnaire, 72.9% reported being in work or in study in 2007. Chronic S. aureus and P. aeruginosa infections were present in 17.3% and 37.7% of patients respectively. B. cepacia was detected in 3.7% in patients at their clinical visit, MRSA in 3.5% and H. influenzae in 7.4%. FEV1 (% predicted) varied between 12.2% and 186.1% (median=73.2%) in patients aged 8 years and older and tended to be higher in younger patients than older patients. Age 8-11yrs median (range) FEV1 (% predicted) = 87% (19.6%-140.8%), 16-19 yrs = 74.4% (15.8%-186.1%), 28-31yrs = 63.7% (12.5%-163.2%), 40+ yrs = 57.8% (13.5%-143.7%). There was little difference between males and females. Among adults, BMI increased with age and ranged between 14.1 and 41.0 (median=21.2) among females, and between 14.9 and 51.1 (median=22.0) among males. Given the ages of patients in the Registry and the mortality distribution of deaths in 2007, it was predicted that half of the current UK CF Registry patients would be expected to live beyond 35.2 years (95% CI: 31.0, 42.6).

Conclusions: This data demonstrates improved survival for UK patients compared to previously quoted figures. Furthermore, the data from Port CF provides a firm basis for monitoring changes in outcomes and performing international comparisons over the coming years. M. 1 ; Yahav, Y. 2 ; Kerem, E. 3 ; Blau, H. 1 ; Rivlin, Y. 4 ; Bentur, L. 5 ; Aviram, M. 6 ; Picard, E. 7 ; Efrati, O. 2 ; Shoseyov, D. 3 ; Livnat, G. 5 ; Mussaffi, H. 1 1. Respiratory Institute, Schneider CMCI, Petach Tikwa, Israel; 2. National CF center, Sheba Medical center, Ramat Gan, Israel; 3. Cystic Fibrosis Center, Hadassah Mount Scopus, Jerusalem, Israel; 4. Cystic Fibrosis Center, Carmel Medical Center, Haifa, Israel; 5. Cystic Fibrosis center, Rambam Medical center, Haifa, Israel; 6. Cystic Fibrosis center, Soroka Medical center, Beer Sheva, Israel; 7. Pulmonary clinic, Sharee Zedek, Jerusalem, Israel Introduction: Life expectancy in CF has improved greatly in recent decades. The establishment of CF databases around the world allowed the CF centers to compare different approaches and to better assess and improve their care. We present the recently established Israeli CF database registry which aims to characterize the genetic and clinical manifestations of the Israeli CF population and to assess the treatment and outcome of these patients.

Epidemiological, genetic and clinical data of CF patients followed at all CF centers (n = 7) in Israel were entered into database registry during the years 2006-7. Findings were compared to the data of the 2005-6 European CF Registry (ECFR).

Results: In 2007, 560 patients were followed in 2007. Data were available for 510 (91%) subjects (males 56%, median age 17y. range 0-58y). Forty-three percent of the patients were >18y (compared to 47% in the ECFR). Thirty-seven percent of the adults were married.

The median age at diagnosis was 0.53y (0-51y). Meconium ileus was the presenting symptom in 14%.

Stop mutations were found in 54% of patients (37% W1282X). DeltaF508, the most common mutation in the Caucasian population worldwide (88% in the ECFR), was found in only 30% of Israeli patients.

Median FEV1 was 83% predicted compared to 75% in the ECFR. Lung function was normal in the first decade (mean FEV1±SD -91±18%), and decreased to 65±23% at 30-34 yrs age group.

Chronic growth of P. aeruginosa in sputum culture was observed in 48% of the patients (similar to the ECFR-43%); 73% of these were multi resistant. Burkholderia cepacia was isolated in 1% of patients (2.5% in the ECFR).

Twenty-six percent of patients were pancreatic sufficient. Median BMI z-score was -0.28. Median weight in CF children was at the 30th percentile. CF adults with pancreatic insufficiency had median BMI of 21.5.

Distal intestinal obstruction syndrome (DIOS) was reported in 42 (9%) in 2006. Three patients with DIOS were pancreatic sufficient.

Eighteen patients (4%) received supplemental gastrostomy feeding. CF related insulin-dependent diabetes was reported in 49 (10%). The prevalence of diabetes in patients >20y was 23%.

Forty-seven percent of patients were treated with inhaled antibiotics and 41% with Pulmozyme (DNAse). A course of IV antibiotics (median duration 16 days) was prescribed to 106 patients (26%) during 2007.

Summary: There are unique demographic and genetic characteristics for the Israeli CF population. Clinical outcome is comparable to large CF centers in Europe. Introduction: Dehydration in cystic fibrosis (CF) patients is an important symptom particularly in hot climates. Dehydration, due to excess loss and/or inadequate intake of salt and fluids, may develop acutely or insidiously (as metabolic alkalosis, especially in infants). Severity varies from mild cases that may pass unnoticed, to severe episodes of hypotonic dehydration that may lead to shock, seizures and death.

Purpose: Our purpose was to investigate the frequency of dehydration among the different clinical presentations of CF in the Greek population and to assess the potential of dehydration as a symptom facilitating the timely diagnosis of CF in this setting.

Methods: During the years 1986 -2008, 521 children were diagnosed having CF in our center (264 boys [50.7%] and 257 girls [49.3%]). Demographic data, the age of patients at the time of diagnosis, and the frequency of the different clinical presentations including dehydration were analyzed.

Results: In a total of 521 newly diagnosed patients, 307 were < 1 year old (59%), 113 were 1 -6 years old (22%), 54 were 6 -14 years old (10%), and 47 were > 14 years old (9%). Among the 521 diagnosed with CF, 152 children had symptoms from the lower respiratory system (29%), 162 children had symptoms from the gastrointestinal system or were malnourished (31%), 116 had dehydration (22%), 68 had meconium ileus (13%), 15 presented with nasal polyps (3%) and 8 with obstructive azoospermia (2%). Among the 116 children diagnosed with CF due to episodes of dehydration (56 boys and 60 girls), 81 were < 1 year old (70%), 27 were 1-6 years old (23%), 7 were 6 -14 years old (6%), and 1 was > 14 years old. In 98 (84%) patients, dehydration occurred during the summer months (from May to September). Of note, 51/116 (44%) children were diagnosed during July (usually the warmest month of the year). In the exceptionally warm years of 1987 and 2007 we noticed a significant rise in the cases of dehydration. More specifically, in 1987, out of 28 newly diagnosed children, 15 present-ed with dehydration (54%). The relevant frequency during 2007 was 33% (11 out of 33 newly diagnosed children presented with dehydration).

Conclusions: Dehydration was the presenting symptom of CF in 22% of patients in our center. Furthermore, whenever environmental changes happen like in 1987 and 2007, dehydration becomes the commonest presentation. The high frequency of dehydration among infants may contribute to the early suspicion and diagnosis of CF, prior to the onset of symptoms from the respiratory and gastrointestinal systems. Considering the global environmental changes, dehydration, as a clinical presentation, will probably be of growing significance in the future. Methods: Starting 16Jul07 all infants born in CA were screened for CF using the CA model which has 4 steps: (1) measuring immunoreactive trypsinogen (IRT) levels in all newborn blood spot specimens, (2) specially designed CA (29-40) CFTR mutation panel testing of specimens with high IRT values (≥62 ng/mL, top 1.5%), (3) CFTR full gene sequence analysis utilizing scanning-sequencing technology of specimens found to have only 1 mutation in Step 2, and (4) referring infants with 2 or more mutations/variants to CF Care Centers (CFCs) for sweat chloride (SC) testing and follow up. Infants with only 1 mutation detected are not referred for sweat testing; their mothers are sent a letter describing the infant's carrier status and offered telephone genetic counseling. CFCs are asked to report all newly diagnosed CF cases not referred to them by the screening program (false negatives (FN)).

Results: Results will be updated to include all observations from the first full two years of screening. As of 15Apr09, 966,604 newborns were screened. In total, 351 newborns had positive CF NBS results: 77 had 2 mutations detected during Step 2 and 274 had 1 mutation detected during Step 2 and 1 or more additional CFTR mutations or variants detected during Step 3. To date, 122 CF cases were successfully detected and 8 had FN screening results (case detection rate = 94%). One FN carried CFTR mutations not on the CA panel and 7 FNs had IRT levels below the cutoff. The false positive rate was low: the ratio of infants referred to CF cases detected was 2.9 : 1 and the ratio of infants referred to infants requiring long term CFC follow up was 1.1 : 1. For the first full year, CF prevalence was 1 in 6,313. CF cases with 2 mutations identified by Step 2 were significantly younger at treatment initiation than CF cases identified by Step 3 (excluding meconium ileus (MI) cases, median ages: 26 vs. 61 days). Most (94%) non-MI cases identified by Step 2 began treatment at age ≤60 days, while only 44% of cases identified by Step 3 began treatment at age ≤60 days. Letters were sent to mothers of 514 infants with only 1 identified CFTR mutation after Step 3, of whom 19% received telephone genetic counseling.

Conclusions: In its first two years, the CA CF NBS model efficiently detected CF cases in a large, multi-ethnic population. Weaknesses of the CA model included a delay in treatment initiation among those requiring Step 3 DNA analysis, a lengthy follow-up period for the majority of infants testing positive during Step 3 in order to determine the significance of CFTR variants detected, and a low participation rate in voluntary State-provided telephone counseling of parents of CF carriers.

Pastore, M.T. 1, 3 ; May, A.E. 2, 3 ; Holtzlander, M. 2, 3 ; Thompson, R.A. 2, 3 ; McCoy, K.S. 2, 3 1. Genetics, Nationwide Children's Hospital, Columbus, OH, USA; 2. Pulmonary Medicine, Nationwide Children's Hospital, Columbus, OH, USA; 3. Pediatrics, The Ohio State University, Columbus, OH, USA Introduction: Newborn screening (NBS) for cystic fibrosis (CF) has been associated with improved nutrition and suggested pulmonary benefits compared with clinical diagnosis. There have also been reports of adverse occurrences, e.g. early colonization with exposure to older colonized CF patients and mixed reviews regarding psychological effects for families. NBS was implemented for CF in Ohio August 30, 2006, through the regional genetics centers and CF Centers.

Methods: Initial screening was by immunoreactive trypsinogen (IRT); those with IRT within the top 5 percent for that day secondarily undergo mutation analysis for the 23 ACMG-recommended common mutations. In April 2008, an additional 19 mutations were added to the screen. Those with at least 1 mutation are determined to be at increased risk of CF. These results are sent to the primary care physician and CF center physicians. Each at-risk infant is scheduled for sweat chloride testing and seen immediately afterward for genetic counseling; infants with positive results are seen by pulmonary physician and treatment started. All are seen in a separate clinic from the established CF population to limit pathogen exposure. Those found to have classic CF are included in our reported results.

Results: As of May 1, 2009, 336 newborns in Central and Southeastern Ohio were identified at-risk. Of these, 286 were unaffected carriers, 5 had elevated IRT, but were unaffected, 16 were seen elsewhere or lost to followup. Twenty-nine were positive by mutation analysis and/or sweat analysis. Of these, 22 have classic CF and 7 non-classic CF. Patient characteristics include: age range 0.5-31.5 months, of the classic CF 14 are homozygous for ∆F508, 5 are ∆F508 compound heterozygotes, 2 were compound heterozygotes not including ∆F508 and 1 was N1303K/unknown. Fourteen had weight/length (wt/lt) percentiles >50, of those lower than 50 the mean wt/lt percentile was 20, with a range of 10-35. All patients have normal vitamin E levels. Sixteen have had IPFT with mean FEV 0.5 percent predicted 96 (range 70-118), 6 have not been tested yet, primarily due to age. Thirteen have had CF pathogens recovered, of these, 3 have CF-affected siblings. Nine have cultured no pathogens to date.

Acceptance of this program by parents and primary care physicians has been good; a formal parent survey is in process.

Problems identified have been primary care physicians who have not followed our protocol and those infants with meconium ileus, where a delay in consultation of the CF team can occur.

Conclusions: A high quality NBS program for CF is possible within a busy CF center, and with appropriate planning, separation of these infants from the colonized CF population is feasible. Early performances regarding weight, vitamin levels, PFT appear excellent. Despite precautions, however, over half of the patients show Gram negative CF pathogens before 3 years of age, though none show a mucoid pathogen. Further education of primary care physicians and non-pulmonary subspecialty colleagues is planned. 

Objective: To support the Newborn Screening (NBS) community worldwide, the Newborn Screening Quality Assurance Program (NSQAP) developed 3 proficiency testing (PT) programs for cystic fibrosis (CF) screening: immunoreactive trypsinogen (IRT), IRT/F508del, and multiple mutation detection. As the program developed to meet screening laboratories' needs, the PT program adjusted the types of materials offered from its inception in 2002.

Methods: Dried-blood spot (DBS) materials for IRT testing were created using a whole blood matrix and IRT enrichment (2002-present) . DBS materials for DNA testing were created using a whole blood matrix enriched with IRT, and transformed lymphoblast cell lines containing 1 or 2 copies of the F508del mutation (2004) (2005) (2006) (2007) (2008) or were prepared from whole blood donated by adult or adolescent CF patients (2007) (2008) . PT specimens were sent to participating laboratories quarterly. The data collected from these programs included test method, clinical assessment, IRT concentration and/or genotype detected. Clinical assessments for IRT were based on whether the specimen was within or outside normal limits. Clinical assessments for DNA were determined by the genotype of the specimen. Laboratory performance was evaluated based on the clinical assessment provided.

Results: In 2009, 139 laboratories participated in the IRT program and 40 laboratories in the DNA program (the IRT/DNA program ended in 2008). IRT and F508del data are based on results reported from 2002-2007 and multiple mutation data are based on results reported from 2007-2008. Using all IRT assessments from both the IRT and IRT/F508del PT programs, the IRT false-negative rate for U.S. and Canadian laboratories combined and for international laboratories was less than 1% (U.S. and Canada: 0.86%, n=1628 results and international: 0.65% n=3409). Correct clinical assessments based on F508del analysis (IRT/F508del program) were reported 96% of the time from U.S. and Canadian laboratories with a 2% sample failure rate (n=258 assessments) and 84% of the time with a 10% sample failure rate from international participants (n=300 assessments). For the multiple mutation detection program using CF donor blood, U.S. and Canadian participants reported 98% correct assessments (n=609 assessments) and international participants reported 94% correct assessments (n=295). Since each NBS laboratory determines the mutation panel that most closely follows their population, there is a wide variety of test panels in use. Therefore, 7% of U.S. and Canadian assessments and 35% of international assessments were not evaluated because 1 or both mutations were not on the laboratory's test panel. Sample failure occurred 0.6% of the time for U.S. and Canadian participants and 2% of the time for international participants.

Conclusions: Data from NSQAP's PT programs for CF screening demonstrates that NBS laboratories worldwide are providing consistent and reliable results. This translates into timely detection and intervention for newborn CF patients. PT programs are a critical component for ensuring quality laboratory performance.

Ms. Driscoll-Dunn was funded by an interagency agreement with the US Department of Energy administered by the Oak Ridge Institute for Science and Education. program for cystic fibrosis (CF). The program uses both immunoreactive trypsinogen and detailed genetic analysis thereby enabling the detection of both novel and less severe variants. Despite the presence of genetic-based diagnosis, guidelines applied by CF Centers call for a diagnosis of CF on the basis of elevated sweat chloride results. The uncertain relationship between genotype and phenotype creates a significant diagnostic and nosologic dilemma for clinicians and families. Thus, the present study was undertaken to test the hypothesis that a correlation exists between genotype and phenotype even upon initial presentation. To test this hypothesis, the CF Center at Stanford evaluated 34 infants referred on the basis of the NBS over an 18 month time interval. Median age at first encounter was 43 days (range 14 to 60 days). All infants underwent a standard assessment that included anthropometrics, sweat chloride (SC) testing, and fecal elastase-1 determination. CFTR mutations were categorized according to their functional class, with classes I, II and III considered severe and classes IV, V mild. Ten infants had a severe genotype (SG), fifteen a mild genotype (MG), and nine infants had mutations which were unclassified (UC). Based on the results of first assessment, five infants were dismissed as not at risk for developing CF disease (all weight >50%ile, SC<15 mmol/L and fecal elastase >350). Of the remaining 29, based on fecal elastase, 20 were classified as pancreatic sufficient (PS), 9 insufficient (PI), and 2 are still pending. All but one infant in the SG group were PI, all but one infant in the MG group were PS and all but one infant in the UC group were PS. For the entire cohort, Median Birth weight percentile was 35%ile, with the lowest value observed in the UC group (25%ile vs 37%ile in SG and 46%ile in MG). These 29 infants were assessed for the degree of malnutrition present at the time of first encounter. At the time of first visit a slight drop in weight percentile to the 30%ile (range 1 to 82) had occurred and weight for length was at the 47%ile. Com-pared to birthweight all genotypic groups demonstrated a slight decline in the median weight percentiles from birth . Weightfor-length (WFL) was already less than adequate only for the SG group (30%ile SG, 48%ile UC, 47%ile MG). The SC values for the entire cohort had a median of 54 mmol/L (range 10-127). Median SC for the SG, UC and MG groups were 116, 84 and 40 respectively. By SC values, infants with SC <30 demonstrated higher birth and first encounter weight percentiles than those with borderline (30-60 mmol/L) or high SC. However, their median weight for length was comparable to that of the infants with high SC (20 %ile vs 30%ile) and infants with borderline SC had the highest WFL (63%ile). This data indicate that genotype phenotype predictions are unreliable in this cohort of newborns. We conclude that genotype phenotype correlations are not absolute in the NBS population. Further, a genetic based diagnosis allows for greater sensitivity in the detection of infants early in the development of nutritional deficiencies regardless of what would be predicted from other parameters such as SC or pancreatic status. Objective: Sweat chloride (SC) testing remains the gold standard diagnostic test for cystic fibrosis (CF) and is the last step in all CF newborn screening (NBS) programs. As a result, the performance and reliability of SC testing impacts the effectiveness of all CF NBS programs. We evaluated SC testing performance and [SC] by CFTR mutation class among infants identified by CF NBS in California (CA).

Methods: Subjects were identified by the CA 4-step model: 1) IRT (≥62 ng/mL), 2) CA-specific 40 CFTR mutation panel, 3) CFTR full gene analysis utilizing scanning-sequencing technology for specimens with only 1 mutation detected in Step 2, and 4) SC testing and follow-up by CF Care Centers (CFCs) for infants with 2 or more mutations/variants. CFCs attempted to repeat SC testing on all infants with initial [SC] ≤59 mEq/L. Number of SC tests conducted and rate of insufficient quantity (QNS) were analyzed for all screen positive infants. Infants' CFTR mutations were categorized according to their class into Groups A (classes I, II, III), B (classes IV, V) or C (Unclassified). Initial and highest [SC] were analyzed by CFTR mutation Group (mutation1/mutation2: A/A, A/B, A/C, B/C, C/C). Infants carrying 1 mutation plus a IVS8-T5 only (N=71), all mutations known to be in cis (N=3), or ≥3 mutations/variants (N=28) were excluded from the analysis by mutation class.

Results: During the first 1.5 years of CF NBS in CA, 302 newborns had positive CF NBS results. Ninety percent (N=273) had at least one SC test performed. Reasons for no SC test conducted included testing deemed unnecessary because of symptoms and genotype, poor infant health, infant death, and parent refused. Among infants with SC results available, 87% had adequate sweat quantity obtained on the first attempt, 8% on the second attempt, 3% required three or more attempts, and 2% are still pending repeat testing. Infants found to have 2 mutations after Step 2 testing were significantly younger at first SC test than infants who had their 2nd mutation/variant detected during Step 3 (median ages: 5 vs. 9 weeks). Infants with initial QNS results were retested at a median 8 weeks after the initial test. Infants with initial sufficient results ≤59 mEq/L were tested again at a median 20 weeks after the initial test. Among screen positive infants meeting the inclusion criteria (N=200), 71 (36%) were classified as A/A, 21 (10%) as A/B, 84 (42%) as A/C, 3 (2%) as B/C and 21 (10%) as C/C. Median initial [SC] (mEq/L) was highest among infants in group A/A (96), followed by groups A/B (31), B/C (21), A/C (16), and C/C (12 The foundation of this work is to report our experience from planning to implementation of newborn screening for CF (NBS CF) in the state of Florida and our findings for the year 2008.

Methods: From the CF newborn screening reports sent to the state of Florida health department by the CF center directors, we extracted mutation data of the patients diagnosed during this year. We also gathered data from the State Lab on the total of patients and specimens processed and the numbers of patients lost to follow-up. We looked at the impact in the change in the cut-off in the IRT top percent in the initial stages of the screening and the elimination of reporting Ultrahigh IRT with no mutation in patients coming from the NICU. We also correlated our rationale for selecting a specific 40 DNA mutation panel and the impact on the number of patients diagnosed with CF.

Results: In 2008, 609 patients were referred to the CF centers in Florida. A total of 41 patients were diagnosed and confirmed with sweat test and or 2-mutation panel. In 8 cases with 2 mutations identified, 2 cases had sweat tests considered possible CF; 3 cases with chloride levels (CL) between 20 to 25 mEq/L and 3 with CL below 16 mEq/L were categorized as uncertain and referred for close follow-up for symptoms. All Ultrahigh IRT blood was sent for DNA analysis. Patients with Ultrahigh IRT and no mutations were referred to the CF centers only if they were not from the Neonatal Intensive Care Unit (NICU) or a second blood sample was tested and high IRT persisted. Ninety percent of the Ultrahigh IRT specimens and no mutations came from patients in the NICU. There were 18 patients with Ultrahigh IRT and no mutations: 16 were negative for CF, one refused testing, and one premature pateint died prior to testing. Ultrahigh IRT and one mutation totaled 15 cases: 9 had negative sweat test, 2 died as premature patients, 1 refused testing, 2 confirmed, one had mecomiun ileus and one positive sweat test. There were 8 cases of Ultrahigh IRT and 2 mutations. The most common disease-causing mutation was ∆F508 (61.3%). Forty-five percent of the CF patients were homozygous ∆F508. A total of 29 parents refused testing and 23 were lost to follow-up or moved out of the state.

From 2007 the DNA testing was done in the top 5% high IRT. In March 2008 this was dropped to 4%, resulting in a decrease of 26% in specimen tested.

Conclusions: Implementation of CF NBS in FL has been very well conducted. The careful planning and collaboration between the Florida health department and the CF centers from the beginning stages including selecting the CFTR panel based on the DNA CF Foundation data from the state of Florida to specific delineation of the contracts between the State and the centers with clear responsibilities and duties and the direct lines of communication between the different parts involved for the tracking of the patients decreases the number of patients lost to follow-up and has established a screening algorithm inclusive despite the diverse population in Florida. The "fail-safe" method of the Ultrahigh IRT in the IRT/DNA model is most likely unnecessary if the correct DNA CFTR panel is selected, eliminating the report to the CF centers of the initial blood samples from the NICUs. This most likely will not decrease the sensitivity for diagnosing patients with CF in the NBS. Results: There were 6 Centers in the LOW group, and the mean open case rate was 4.7% (range=1.9-8.4%). There were 6 Centers in the HIGH group, with a mean open case rate of 17.7% (range=12.5-21.6%). The results of the survey are summarized in the Table. More primary care practitioner (PCP) involvement, and upstate location were significantly associated with LOW centers. LOW centers also had more FTE devoted to NBS per 100 referrals, more hospital support for NBS, and a lower percentage of unknown PCP, although these differences were not statistically significant compared to HIGH centers. We are presently investigating other factors, such as practice setting, that may also be associated with a high open case rate.

Conclusions: There is wide variation in the open case rate for CF NBS in NY. Our study identifies factors associated with a high open case rate and potential areas for quality improvement to reduce the open case rate. (CFRD) has been shown to lead to poor nutritional and pulmonary outcomes. As a result, early diagnosis and treatment of CFRD has the potential to enhance patient outcomes and improve survival. The Johns Hopkins Pediatric Cystic Fibrosis Center team identified three primary outcomes, related to CFRD, to focus on for our quality improvement (QI) effort: screening, treatment, and patient education.

Methods: Our team has bi-weekly multidisciplinary QI meetings to discuss the development, implementation, and evaluation of quality improvement initiatives at our center. One goal was to design standardized CFRD screening and treatment protocols for both inpatients and outpatients which incorporated interpretation of oral glucose tolerance test (OGTT) results. The new protocols have a lower threshold for treating hyperglycemia than was previously used in our clinic and are designed to trigger early intervention. Patients with abnormal testing are given a glucometer to do home monitoring and are referred to a pediatric endocrinologist. Patient education designed specifically for individuals with CFRD, as well as impaired glucose tolerance, was identified as an area in need of improvement at our institution. An educational flipchart and handouts were developed for use in the outpatient clinic and inpatient units. Each patient identified as having CFRD or impaired glucose tolerance through the screening process receives indi-vidualized education and a "CFRD bag" with educational materials, a glucometer, and a carbohydrate counting book.

Discussion: The development and implementation of standardized protocols and education materials for the screening and treatment of CFRD will allow for early identification and treatment of patients at risk. This approach has the potential to improve the pulmonary and nutritional outcomes of our patients. This program is currently being implemented at our CF center. Future goals are to track patients identified through this screening process and evaluate the usefulness of the educational flipchart and handouts. Foundation recommend an annual assessment and education review by a respiratory therapist, dietician, and social worker. The UAB/CHS center has historically compiled delivery of this care with nursing review, laboratory and radiologic assessment at an "annual visit" (AV). This resulted in extensive, 4-hour sessions for patients, families, and team members. We began an interdisciplinary initiative aimed to redistribute this burden throughout the year. The goals of this effort included: improving clinic efficiency; increasing patient, family and team satisfaction with education; and increasing patient and family comprehension of education.

Methods: In 2007, the UAB/CHS center developed a rotating schedule of quarterly annual education reviews by each discipline and routine diagnostic testing was scheduled to occur during summer months. The outcomes of these changes were measured by team, patient and family satisfaction surveys. Clinic time cycles were conducted to evaluate total clinic time and productive clinic time.

Results: Team satisfaction surveys revealed that all team members except the dieticians thought the annual review was necessary. Dieticians see the patients at each visit and have increased patient access. Care team members thought the redistribution of care to annual reviews improved education and clinic efficiency (100%), contributed to patient care (100%), and improved contribution to interdisciplinary care (88%).

Patient surveys comparing the new annual review to the previous AV model indicated that the annual review was helpful (81%), met their educational needs (88%), that education was good or excellent (83%), and noted improvement in education organization (63%).

The 2008 time cycles as recorded by patients showed clinic visit time averaged 144 minutes, with 81 minutes as productive time with a team member (productivity = 56%). In 2007, total clinic visit time averaged 105 min, with 66 minutes as productive time (productivity = 62%), revealing an expected increase in productive clinical time but a trend toward decreased efficiency. This most likely represents a period of adjustment to the new model of care delivery. Despite the increased time burden in the clinic, 50% of patients/families surveyed felt that the redistribution of care was a significant improvement in delivery of care and educational experience.

Conclusions: We conclude that a balanced distribution of yearly diagnostic tests, education and review was preferred over a single comprehensive care review by patients, families and CF care team members. Despite a trend toward diminished efficiency, half of the patients/families surveyed perceived a significant improvement in care delivery. In an effort to decrease non-productive clinic time we have implemented a clinic communication sheet and flowchart and restructured physician and nurse roles to improve efficiency of all team members. Additionally, in the summer of 2009, the UAB/CHS center will expand the physical clinic space to improve contact isolation and reduce patient load in clinic. A time cycle and surveys will be repeated after these changes. Background: Cystic fibrosis (CF) clinics involve patient evaluation by a multi-disciplinary team consisting of physicians, nurses and nurse specialist, the respiratory therapist (RT), dietitian (RD) and social worker. Running clinics efficiently becomes a challenge with increasing patient volume, volume of clinical information, and increased complexity of available therapies. Therefore, we developed an algorithm for pre-clinic screening that could become part of the patient's electronic medical record, summarizing key aspects of patient outcomes and therapies.

Hypothesis: An EMR tool to help the cystic fibrosis caregiver teams to rapidly assess disease status and current therapies improves quality of care delivered.

Methods: 1. An Excel file was first drafted to include dates of routine and annual visits; results of sputum culture, FEV1 (% predicted), pertinent pulmonary medications (rhDNAse, hypertonic saline, azithromycin, inhaled antibiotics (tobramycin, colistimethate sodium, aztreonam), nutritional parameters (weight, height, BMI percentile, vitamin deficiencies), and recommended evaluations (OGTT, DEXA scan) and was approved and reviewed by all the CF team members.

2. Clinical Informatics prepared a pre-clinic screen including all these parameters in a format compatible with the outpatient electronic medical record (Allscripts Touchworks ® ) to allow tracking over time.

3. Two weeks before the anticipated clinic visit, the RD completes nutritional parameters and the RT completes pulmonary function results. The CF nurse specialist then completes the rest of the screen.

4. The entire CF care team reviews the completed document for each patient at a clinic conference one week prior to the scheduled clinic visit to determine needed evaluations, or new therapies that may be indicated.

Results: We implemented the CF pre-clinic screening routinely for all our patients in the beginning of 2008.

Outcomes were compared between quarters and are presented in the Table 1 below.

The percentage of eligible patients completing the diabetes screen increased from less than 20 in the first quarter to 68% in the third quarter.

Conclusion: Effective pre-clinic screen greatly increases the quality of care and is key in improving outcomes in a busy multi-disciplinary CF clinic. Background: Infection control measures are essential in order to prevent the spread of bacteria between people with CF. Although it is assumed that people with CF have extensive knowledge about the importance of infection control measures and what constitutes risky behaviors, the amount and depth of patients' knowledge on this topic is variable. A recent outbreak of Burkholderia cepacia complex (BCC) at our centre prompted our team to question the knowledge of pathogen transmission in our patient population.

Purpose: To assess current infection control understanding and beliefs among people with CF at our center.

Methods: Over a 4-week period, patients attending the Adult CF program at St. Michael's outpatient clinic were invited to participate in the study. Subjects anonymously completed the KAP Survey for CF Patients and Families/Infection Control in CF in order to assess their knowledge of infection control. This survey included questions regarding the risk of having contact with others with CF, hand hygiene and nebulizer cleaning.

Results: Sixty-nine patients that attended the outpatient clinic during the study period completed the survey. The mean age of subjects was 35 y. Half of the respondents had been hospitalized in the last year. Regarding contact with other CF patients, the majority of subjects (88%) agreed that germs are spread from one person with CF to another. However, 31% indicated that socialization while in hospital was acceptable. Although 79% agreed that people with CF should be at least 3 ft apart, 15% of respondents felt it was safe to be within this distance if neither person were coughing. Sixty-two percent stated they were told by their CF team to avoid close contact with people with CF. Even though 1/3 of subjects stated the CF team had not specifically reviewed hand hygiene, all subjects believed that cleaning hands frequently reduced the chance of getting infections. Sixty-four percent of subjects felt it was very easy to clean their nebulizer; however 36% of patients surveyed cleaned it less frequently than recommended. Patients felt the hospital lacked appropriate facilities to properly clean their nebulizer equipment. One-third of patients stated they had not received information from the CF team regarding the cleaning of nebulizers. Several patients commented that a standard protocol for cleaning nebulizers would be useful. Subjects indicated that they would appreciate frequent updates (both oral and written) on infection control issues important to individuals with CF.

Conclusions: Although our patient population had a good understanding of infection control, several gaps in knowledge were identified. Furthermore, patients stated they had not received important infection control information from the CF team. As a result of this survey, our centre has implemented several quality improvement changes to our practice including the development of a detailed protocol outlining the process and frequency of nebulizer cleaning. Also, an infection control education checklist is being developed with input from our patient/family advisory board. This education tool will be reviewed with all individuals in the outpatient and inpatient settings. Background: Pulmonary exacerbations are a frequent cause of hospitalization in CF and a major contributor to the long-term deterioration in lung function. Hospitalizations for pulmonary exacerbations involve a number of activities and treatments that can be difficult to coordinate. We developed a daily patient schedule to ensure that all treatments are performed at appropriate frequency and intervals in order to make hospital stays more efficient and productive, with the expectation that this would lead to better outcomes.

Methods: A multidisciplinary steering committee was formed and included a family council representative to help maintain a patient-and family-centered approach. We designated a group of core CF nurses who were given the opportunity to attend a multidisciplinary half-day educational seminar presented by CF Center personnel called "CF Academy." Core nurses signed a contract stating they would fully support the scheduling and incentive initiatives. On admission, patients and families were provided with an education packet detailing the goals of the program, and invited to participate. They were asked to sign a contract stating they understood the program and would support its activities. The core nurse met with the family to discuss the program and develop a schedule of all activities, and then communicated the schedule to the interdisciplinary team. Patients were given tokens as incentive for (1) on time wake-up and availability for morning assessments and meds; (2) consuming all meals and dietary supplements; and (3) observing nighttime curfew. They also received tokens from other disciplines for participating with airway clearance therapies, exercise, and calorie counts. We monitored program effectiveness and acceptance by surveying patients at discharge, and adjusted our approach based on this input.

Results: Of the first 19 patients who participated, 9 were female, and their average age was 16.3 years (range 10-19 years) and average LOS was 14.7 days (range 8-30 days). Of a maximum of 5 tokens/day that could be awarded by the nursing team, the average given was 3.9/day, with a range of 2.5-5. Results of our satisfaction survey showed that 82% of patients agreed or strongly agreed that the program made the hospitalization more effective. Our lowest score was received in regards to availability of enzymes when needed, and this issue continues to be addressed. It is too soon to evaluate the effect of this program on changes in BMI or FEV1.

Conclusions: We attempted to improve care for CF pulmonary exacerbation by educating patients on the goals of the hospitalization, establishing a schedule to help them anticipate when treatments will be provided, and encouraging active participation in these treatments. The patient and parent response to this program have been positive and we believe that it will help us to improve the outcomes of hospitalization. We plan to apply the program to all admissions, and begin a new phase involving the identification and remediation of gaps in CF knowledge among patients and families. Background: Cystic fibrosis (CF) is a genetic disease associated with an accumulation of tenacious mucous causing airway obstruction. Chest physiotherapy (CPT) facilitates the removal of these secretions. Due to the chronic nature of the disease and the labor-intensive therapy, non-adherence to CPT is well documented in the literature. Patients' knowledge and participation in regular CPT may be increased with education and recommendations provided by the CF physiotherapist.

Purpose: The objective of this quality improvement initiative was to develop a strategy to identify high priority patients in need of physiotherapy consultation and to increase the efficiency with which patients are seen in the ambulatory clinic.

Methods: The CF clinic at St Michael's Hospital (SMH) in Toronto, Canada follows over 400 adult patients. Two full-time physiotherapists are responsible for a 15 bed inpatient respirology unit in addition to the ambulatory clinics. Outpatient physiotherapy consultations are provided on an asneeded basis. Patients requiring physiotherapy services were traditionally identified during clinic by the CF nurse who documents the type of physiotherapy used, duration, and frequency. However, this process was not standardized and physiotherapists had no advance notice of outpatient consultations. Thus, criteria for "adequate" CPT were established by the physiotherapists based on the frequency and duration of patients' CPT regimen. Patients were then categorized into "high," "moderate," and "low" priority for physiotherapy consultation using a standard algorithm during the weekly chart review meeting. Patients identified as "high" priority were flagged and highlighted on the clinic list prior to their next appointment. Our shortterm goal was to provide outpatient physiotherapy consultations to 100% of individuals identified as "high" priority and the long-term goal is to track the proportion of patients doing "adequate" physiotherapy.

Results: Preliminary results showed that over the last five months, 106/252 (42%) "high" priority patients have been identified. To date, 12 "high" priority patients have returned to the ambulatory clinic. Ten (83%) received physiotherapy consultations during their follow-up clinic visit. During this consultation, patients received general education on available physiotherapy techniques, prescriptions for specific airway clearance exercises/devices and strategies to incorporate CPT into their daily routine. Two patients were not seen because of caseload constraints. At baseline, 44% of patients were classified as doing "adequate" physiotherapy. We will monitor this every six months and anticipate having updated data by the summer.

Conclusions: By using this strategy, we have identified a large number of CF patients with acute physiotherapy needs. This has allowed us to anticipate the need for outpatient physiotherapy consultations in advance so that the physiotherapists can accommodate these patients in their day. Also, increasing the outpatient physiotherapy presence provides an excellent opportunity to emphasize CPT techniques and increase patients' knowledge which we hope will translate into improved adherence.

Spoonhower, K.; Kotha, K.; Salvatore, A.; Kraynack, N.C. Akron Children's Hospital, Akron, OH, USA Background: Studies of children with cystic fibrosis (CF) demonstrate that indices of growth and nutrition correlate positively with lung function. Although many factors affect growth, aggressive nutritional management may lead to improved linear growth, ultimately resulting in improved lung function and survival for CF patients. Beginning in mid-2006 the Lewis H. Walker CF Center implemented a comprehensive program to improve the nutritional status of pediatric patients with CF that included the use of a standardized Nutritional Assessment Score (NAS). Although the center median BMI percentile increased following consistent use of the NAS (Phillips et al. Pediatr Pulmonol 2008;S31:408), it was unclear if patients were making gains in linear growth. Therefore this study examined linear growth velocity in pediatric patients following implementation of the NAS.

Methods: A retrospective chart review was conducted of all CF patients age 3-18 years old from 2006-2008. Patient height, weight, BMI, age, gender, bacterial colonization, baseline FEV1 (if available) chronic therapies, and co-morbid conditions were recorded. Growth velocity (cm/year) was calculated and recorded for four quarters prior to the standardization and use of the NAS, and for four quarters following standard use of the NAS. Growth velocity curves were compared pre-and post-NAS use. Equality of growth velocity slopes and intercepts in different risk groups were tested by Mixed Model analysis using SAS Proc Mixed (version 9.1). A level of p=0.05 was used to indicate statistical significance.

Results: Sixty-nine patients had data available for the time period described above. Overall, there was no significant difference in linear growth velocity after the implementation of the NAS. However, there was a brief increase in growth velocity shortly after implementation of the NAS that was not maintained. There were differences in the linear growth velocity of CF patients with allergic rhinitis (p<0.04), osteopenia (p<0.03), and although it did not reach statistical significance, differences in the growth velocity of CF patients with CF-related diabetes (CFRD) were striking (P<0.09). There were no differences in growth velocity based on bacterial colonization, pancreatic sufficiency, and other co-morbidities.

Conclusion: Although linear growth velocity briefly increased after implementing the NAS, the increase was not sustained. In addition, patients with allergic rhinitis, osteopenia, and to a lesser extent, CFRD had lower rates of growth overall. Interestingly, pancreatic sufficiency did not predict improved growth velocity. This suggests that improving linear growth may require targeting problems other than malabsorption and caloric intake. Further studies investigating the reasons underlying reduced growth velocity with co-morbidities such as CFRD are needed.

Spoonhower, K.; Hixon, K.; Salvatore, A.; Kraynack, N.C. Akron Children's Hospital, Akron, OH, USA Background: Although pulmonary exacerbations (PEx) contribute to significant morbidity and decline in lung function for patients with cystic fibrosis (CF), no standard definition exists. The Lewis H. Walker CF Center developed a Pulmonary Exacerbation Score (PES) to uniformly identify patients experiencing a PEx, which resulted in an overall increase in the median FEV1 in all pediatric age groups at our center (Kraynack, 2009) . Although the PES identifies a PEx, it is not known if it reflects changes in clinical status. Therefore, we investigated if the PES reflects acute changes in a patient's clinical status including both deterioration and improvement or resolution of a PEx at the end of therapy.

Methods: A retrospective chart review was conducted of all CF center patients aged 6-18 years, treated for PEx from January 2007 to January of 2008. For each PEx, the PES both before and after treatment were recorded. Independent variables collected were as follows: treatment duration, days to follow-up, route of treatment (intravenous versus oral), co-morbidities, and bacterial colonization. Differences in pre-and post-therapy PES scores were calculated and tested by paired t-tests. To test for relationships between the calculated difference of "Pre" and "Post" PES score and independent variables, correlations were estimated. Multiple linear regression analyses were used to construct the best model for predicting the magnitude of difference in the Pre and Post PES score. For this purpose, treatment duration, days to follow-up, route of treatment, co-morbidities and bacterial colonization were tested. P<0.05 was considered significant.

Results: The mean change in PES from start of antibiotic treatment to the end of antibiotic treatment was -3.1 (p<0.0001). The mean difference in PES pre-and post-treatment with intravenous antibiotics was significantly higher (-6.1) compared to the difference in PES in those prescribed oral antibiotics (-2.1) . Variables that significantly predicted a change in PES were days to follow-up (p<0.017), the presence of allergic rhinitis (p<0.009), and route of antibiotic administration (p<0.0002). In addition, there was a trend toward treatment duration predicting change in PES, although it was not statistically significant (p<0.06).

Conclusions: The PES does decrease after therapy for a pulmonary exacerbation, thereby reflecting an acute change in a patient's clinical status. Therefore, the PES can not only be used to identify patients experiencing a PEx, but can also be used to monitor response to treatment status and guide future therapy. Furthermore, additional variables such as route of antibiotic administration and treatment duration may also be used to further standardize therapy such as providing thresholds for the recommended use of intravenous antibiotics or guide treatment duration. The Cystic Fibrosis Foundation (CFF) has developed guideline recommendations to establish standardized cystic fibrosis (CF) management, which has been attributed with improved health outcomes for patients. Only two previous studies have examined whether adherence to the CFF guidelines for patient care is linked to improve health outcomes (Johnson 2003; Padman 2006) . The purpose of this project was to link care center adherence to guideline-based care to center-level health outcomes as measured by mean FEV1 and percent of patients with pulmonary exacerbations.

Methods: This is a cross-sectional study to examine the association between care centers' adherence to clinical care guidelines and center-level patient health outcomes, using data from the CFF Patient Registry (CFFPR) for the year 2007. Measures of guideline-based care were abstracted from the CFFPR as percentage of patients meeting the following criteria in each care center: 1) seen by a dietitian when BMI < 50th percentile, or <22 for female or <23 for male, 2) tobramycin use in P. aeruginosa + patients >= 6 years, 3) dornase alpha use in patients >= 6 years, 4) hypertonic saline use in patients >= 6 years, 5) 4 + clinic visits, 6) 2 + PFTs a year, 7) 1+ culture a year, 8) glucose screening in non-diabetes patients >= 14 years,9) influenza immunization done in patients > 6 months, 10) liver enzymes documented as done. Linear regression models were used to predict health outcomes by all indicators of care centers' adherence to guideline-based care.

Results: Preliminary results indicated that pediatric care centers who had higher mean FEV1 % had a higher percentage of patients who met the guidelines for at least 4 visits a year (p =0.06) and at least 2 PFTs a year (p = 0.05). However, pediatric care centers who had a higher percentage of patients with an acute exacerbation had more patients meeting guidelines for clinic visits (p = 0.01), prescriptions of pulmozyme (p = 0.02) and hypertonic saline (p = 0.05). Similarly, adult care centers who had a higher percentage of adults with an acute exacerbation had more patients meeting the criteria for 4 clinic visits (p = 0.001) and one respiratory culture (p = 0.04).

Conclusions: Results indicated that pediatric centers who met certain guidelines for care had improved lung function. However, an inverse relation was found for both pediatric and adult centers for acute exacerbations, such that centers that had a higher percentage of patients meeting guidelines for care had more patients who had an acute exacerbation. One explanation for this finding is that care centers that see their patients more often and have more respiratory cultures may be more aggressive with antibiotic therapies to prevent lung function decline. These results are similar to earlier findings by Johnson (2003) such that care centers with higher mean FEV1 % values had more frequent clinic visits, PFTs, and respiratory cultures and were more aggressive with IV antibiotics to treat pulmonary exacerbations. Future analyses will include adjusting for case-mix using patient and disease characteristics for each center.

Supported by the Cystic Fibrosis Foundation. The high cost of primary research requires funding agencies to set priorities. Clinical experts are in the best position to identify areas where our ability to make healthcare decisions is limited by a lack of evidence. This is part of the development of evidence-based guidelines, however, the research gaps thus identified are not explicitly used in setting research agendas. Our objective was to pilot test a method to systematically identify gaps in evidence.

Methods: We reviewed all evidence-based guidelines developed by the Cystic Fibrosis Foundation. From each guideline we abstracted a) topics for which recommendations were not made because no evidence was available, because the evidence was insufficient, or because the evidence was conflicting; and b) questions specifically identified by the guidelines committees as needing further research. We classified topics by type of patient addressed and type of question.

Results: We reviewed five guidelines addressing the following: nutrition-related management, chronic medications for maintenance of lung health, airway clearance therapies, treatment of acute pulmonary exacerbations, and care of infants diagnosed with cystic fibrosis. We identified 34 questions (mean 6.6 per guideline). The majority of these were topics where recommendations were not made due to insufficient evidence (n=20). Seven topics were specific to infants. The severity of disease considered was not noted in the majority of questions (n=23). The question type was classified as comparative effectiveness (n=18), integration into practice (n=6) and long-term effects (n=10).

Conclusions: We identified and classified 34 questions. The classification may help to determine research needs. For example, questions of comparative effectiveness may best be addressed by clinical trials while determining the long-term effects of an intervention may be more appropriately addressed by prospective observational studies. The process of developing evidence-based guidelines often exposes areas where we lack evidence to guide practice. Systematically identifying these gaps in evidence provides the opportunity to efficiently direct scarce resources. To determine the accessibility of callers to the Cystic Fibrosis (CF) Centers and their ability to schedule a sweat test for a newborn with a possible diagnosis of CF, and to compare these results with the accessibility to general pediatric practices.

Methods: Contact information on 158 CF centers and affiliates was obtained through the Cystic Fibrosis Foundation (CFF). Avoiding holiday weeks, we called each center twice on Wednesdays or Thursdays; results were averaged for the two calls. Telephone numbers for the first call came from the CF Center Directory provided by the CFF and the second utilized the online directory from the CFF website that can be accessed by anyone. When there was inconsistency among the phone numbers on the two directories, the original number was called again to obtain replication. Using a scripted telephone survey, various data were collected and analyzed, name-ly: length of call, hold time, number of voice prompts, language option, number of times transferred, alternative telephone number provided, if the telephone was answered by a person and the date and time of the next available sweat test for a newborn. The telephone numbers for multi-physician pediatric practices were selected from yellowpages.com by matching to the CF centers' area code and zip code. Call outcomes resulted in the following categories:

A-Successfully completed with one call: The center was able to provide a time and date for a sweat test.

B-Successfully completed with multiple calls: The center was not able to give both a time and date for a sweat test but provided another telephone number that eventually resulted in a successful call.

C-Dead End with multiple calls: The center was not able to give both a time and date for a sweat test but provided another telephone number that resulted in a dead end.

D-Dead End with no response: The center's telephone was not answered and/or there was no answering service or the answering service did not make it clear if the telephone number was associated with the CF Center.

Results: Defining success as category A or B, we found that only 30% of centers were successfully contacted on both rounds of calls, while there was a dead end result on both calls with 27% of the centers. Calls to the remaining 43% of the centers were unsuccessful on at least one approach. The average time spent on the telephone was 2 minutes and 32 seconds. The average time spent on hold was 1 minute and 42 seconds. There was an average wait of 4.85 days to have a sweat test done, but delays as long as 35 days were observed. In contrast, the calls to general pediatricians' offices have resulted in 92% with A ratings, calls being answered by a person in 7 seconds on average, and appointments made within one week in 1 minute and 35 seconds.

Conclusion: Substantial inconsistency was found in access to scheduling a sweat test in CF centers and affiliates both across the nation and within the contact information disseminated by the CFF. These results suggest that parents may often be challenged in their efforts to schedule sweat tests.

Supported by NIH Grant R01DK34108.

The following specific aims will be added in a stepwise fashion: (1) decreasing barriers to adherence in CF patients and families by 50%; (2) improving adherence to airway therapies (RT or ACT) by 35% and to medications (MEDS) by 25%. The global aim was presented to the CCCFC Patient Family Advisory Board for approval. Plan-Do-Study-Act (PDSA) methodology was applied repeatedly to develop changes to care, monitoring tools, and track outcome measures. P: Evaluate current patient-family adherence; assess EBM recommendations -review existing adherence literature and available adherence tools D: Survey sample population; survey barriers to adherence; introduce goals for self-management; develop algorithms for sick plan, nutrition, RT, ACT, and MEDS; develop tools to integrate algorithms into clinical care S: Track change secondary to adherence interventions; reassess number of missed treatments; survey patient-family for effect; assess medication refill history A: Assess progress along self-management continuum; develop educational material for families to operationalize algorithms.

Results: Families were surveyed to determine which area they felt the greatest barrier to delivery of care: MEDS, sick plan, RT and ACT, or nutrition. Families were then asked what it was in that area that most impeded care. Patients were asked to indicate anonymously how many MEDS, RTs, and ACTs they missed in the previous week. There were 40 respondents for MEDS, 33 for RT, and 43 for ACT. At least 50% estimated or admitted missing a MED, 39% missed an RT, and 72% missed an ACT. Forty-six percent said ACT was the hardest treatment to complete on a regular basis, 16% responded nutrition and sick plan were hardest, 14% said RT, and 8% regular MEDS. Of those who felt sick plan, RT, ACT or nutrition were hardest 65%, 50%, 62%, and 7% respectively noted "it takes too long" is the greatest barrier; for those who responded nutrition, regular MED, and sick plan were hardest 53%, 25%, and 5% respectively said disliking "taste" or "feel" as the greatest barrier to adherence. Algorithms to improve time management in RT/ACT, nutrition, MEDS, and use of sick plan were developed. Tools to track outcomes were incorporated into clinic assessments. Runcharts for response to adherence QI in the 6 to 10 year old age range including frequency of missed medication, medication refill history, relative change in barriers to adherence, and achievement of goals for self-management will be presented.

Conclusion: CCCFC's adherence is consistent with previous reports regarding ACTs, MEDS, and nutrition. Given, however, the unsurprising prevalence of the finding "it takes too long" or "not enough time" there is a surprising lack of EBM or QI addressing time management and adherence interventions in the CF or other chronic illness population.

Weiland, J.; Taylor, J.; Browning, G.; Chima, A.; Dressman, K.; Acton, J. Pulmonary Medicine, Cincinnati Children's Hospital, Cincinnati, OH, USA Self management is a process which individuals with chronic illness engage in to manage medical, emotional and functional health. Self management support utilizes motivational interviewing, problem solving and shared decision making to assist patients in making changes that will have a positive influence on health outcomes. Ultimately, the goal of self management support is to help patients learn to develop behaviors that will optimize their health status and quality of life.

At Cincinnati Children's Hospital, a multidisciplinary team of CF care providers attended six self management support training sessions, led by local experts. Participants were trained in basic motivational interviewing techniques and learned how to assess readiness to change and help patients set goals and problem solve barriers to success. A self management bundle, consisting of a readiness assessment, progress tool and goal setting tool, had been developed by the experts leading the training course and was provided to the team members.

Following the course the CF care team staff, utilizing PDSA cycles, developed a process to successfully implement self management support in the clinic and inpatient settings. Initial PDSA cycles focused on simply finding a process to get the bundles to the patient and monitoring length of time to complete each portion of the bundle. To date, multiple small tests of change have resulted in the development of successful processes in both set-tings (clinic and inpatient) to support patients in assessing readiness to change and, when ready, setting goals. We have been successful in devising a process to communicate status of readiness and goals set with the entire CF team prior to clinic visits and hospital discharges.

Currently, the team has focused on a process measure to evaluate success. Completion of the self management bundle by eligible patients is monitored on a monthly basis. Data collection began in January 2008 when the team began its work. With an initial focus on inpatients, a captive population, bundles were completed 100% of the time by eligible patients (eligible at that time was any patient 15 -21 years of age). With implementation of a process in the clinic setting in March 2008, results fell to 75% overall and in April 2008 to 29% for completion of the bundle in eligible patients. Following examination of the process and several PDSA cycles, completion of bundles among eligible patients (which was expanded to include all patients with CF seeing a particular physician in clinic and all 15 -21 year old inpatients) rose to 100% in May 2008.

Self management support is critical in helping patients achieve optimal health outcomes. The first step is development of a process to consistently provide this support to patients and families in CF clinic and the hospital setting. To date, our team has been successful in designing a process to implement self management support in the clinic and inpatient settings. As we become more proficient at this process, the focus will shift from a process measure to how well patients are meeting their goals or moving toward readiness to set goals. Ultimately, we will observe and measure improvement in health and quality of life outcomes. The Cystic Fibrosis (CF) Foundation published evidence-based recommendations for nutrition goals in 2005 stating that a BMI > 50% is associated with lower risk of morbidity and mortality. Our CF Center Quality Improvement efforts focusing on improving nutrition outcomes began in 2002. The initial focus was to classify each patient's nutritional status at every encounter and address barriers to achieving desired BMI. This systematic approach led to improvements in our Center's median BMI percentile from 33.2% in 2002 to 52.4% in 2008. Encouraged by these results, the team was eager to take the next steps. We enhanced our algorithm that has become our center's Nutrition Guideline, created a tool to track interventions over time (Nutrition Tracker), and increased the productivity of our weekly patient care planning meeting (Chart Conference).

The Nutrition Guideline represents the complex challenges that can affect weight gain for individuals with CF and is divided into 6 categories: caloric intake, absorption, Cystic Fibrosis-Related Diabetes, gastroesophageal reflux, psychosocial barriers, and enteral tube feedings. Each category follows an algorithm detailing assessment, medical management and self-management, and ongoing re-assessment. A change in one category often requires adjustments in other categories. A key component of the Nutrition Guideline is assessing the patient's actual weight gain since their last encounter compared to their monthly weight goal. We discovered our patients and families were often unaware of their short-term and long-term weight goals. Together with the patient and family, we now set individualized monthly weight goals based on age, gender and catch-up needs for the patient. If these goals are met, no changes in management are necessary unless the patient/family desires a change. If the interval goals are not achieved, the Nutrition Guideline facilitates the next intervention.

The Nutrition Tracker is a patient-specific flow sheet that is updated with every patient encounter by the Dietitian and coordinates with the Nutrition Guideline. It tracks anthropometric measurements and weight goals as well as specific information for each of the intervention categories. It also includes the plan for reassessment and next steps developed with the patient and family. The Nutrition Tracker is instrumental in effectively coordinating nutrition care by documenting a detailed history of interventions and the patient's response to those interventions.

All members of the multi-disciplinary CF team meet weekly (Chart Conference) to review patients scheduled to be seen in clinic. Prior to using the Nutrition Tracker, discussion of patient management was inefficient and ineffective, depending on memory. With the use of the Tracker, the dietitian is now prepared to present anthropometric trends, current and previous interventions, as well as goals and plans established with the patient and family. Sharing this information across the multidisciplinary team helps identify other concerns and educational opportunities in supporting the complex comprehensive care required by individuals with CF. Objective: Improving quality of life and partnership between health care team and patients/families have driven the efforts of the Cystic Fibrosis (CF) Foundation to organize initiatives such as Learning and Leadership Collaborative (LLC). Sharing this belief, our CF Center has taken part in LLC-VI. Our global aim: improve pulmonary function. Our specific aim: increase mean FEV1 by 10%. An individualized written action plan will be implemented to meet the needs of every patient. We anticipate that by increasing the mean FEV1 by 10%, we will be above national mean FEV1 by one year.

Methods: Individualized action plan was created with input from CF Team and patients/families. Action plan was given to every patient at each visit. Our center has 70 patients: 46 pediatric patients (<18 years) and 24 adult patients (>18 years). Improvement tools such as Plan-Do-Study-Act (PDSA) cycles were used in implementing small changes: revisions of the action plan, reorganization of charts, restructuring of team meetings, streamlining clinic flow, and enhancing communication with patients/families. Port CF was utilized to track outcomes.

Results: Mean FEV1 in our pediatric patients increased: 80% (2007) to 87% (2008) . Mean FEV1 in our adult patients decreased: 61% (2007) to 51% (2008) . Mean FEV1 per quarter showed the pediatric population with improving trend; the adult population with fluctuating trend (see graph). There was a significant increase in adherence to care guidelines for pediatrics, 33% (2007) to 68% (2008), but a significant decrease for adults, 68% (2007) to 48% (2008) .

The following were implemented: creation of CF PAC (Parent Advisory Council); award of two grants: PACE (Program for Adult Care Excellence) and Quality Improvement Grant; Family Education Day, with emphasis on adherence/partnership.

The Action Plan has demonstrated efficacy through improvement in mean FEV1 in our pediatric population. However, due to the small number of our adult population, outcomes were easily impacted by sick episodes. There was a significant improvement in adherence to care guidelines in pediatric patients, but remained a challenge in adults.

Future Plans: Our center will focus on transition to Adult CF care. Our adult physician has joined through the PACE grant. Our small number of adults is a major factor affecting our data. It would be best to continue improvement on adherence to care guidelines, and take on the challenge of improving mean FEV1 in both our pediatric and adult patients to reach the national goal. CF PAC with the involvement of patients/families will play a larger role in educational and team effort events. Aminoglycoside antibiotics are known to be toxic to the inner ear. However, they continue to have a leading role in the treatment of certain infections and in the treatment of pulmonary exacerbations in patients with cystic fibrosis (CF). While there is ample evidence in the literature of ototoxicity, and there are established protocols for monitoring the effects of toxic agents on hearing, less is known about the prevalence of aminoglycoside vestibular toxicity among CF patients, and there are no commonly accepted protocols for monitoring vestibular system function.

Because of our concern about quality of life issues in CF patients with apparent vestibular ototoxicity who were referred to the Vestibular Testing Center (VTC) at the University of Michigan, we initiated a quality improvement clinical protocol with our pediatric CF patients. The purpose of this ongoing investigation is to determine the prevalence of hearing loss and vestibular system involvement among patients with CF who receive intravenous aminoglycosides. A secondary aim is to determine whether incremental changes in auditory and vestibular function can be detected with subsequent courses of antibiotics.

To date, we have completed vestibular testing on 55 patients with CF ranging in age from 10-56 years. Of the sample, using our clinical criteria, 13 have bilateral vestibular loss, 9 have a unilateral vestibular loss, and 19 have non-lateralizing evidence of peripheral vestibular system involvement. We have also documented change over time with repeat testing. Specifically, we have seen change from normal caloric results to a unilateral vestibular loss, from non-lateralizing findings to a bilateral loss, and from a mild bilateral loss to severe bilateral loss.

Interestingly, 13 individuals in the group have documented hearing loss, and only six have both normal hearing and normal vestibular test results. Of note, hearing was not evaluated in 10 patients. It is clear from our limited data that it is important to monitor vestibular system function whenever potentially toxic agents are used. While monitoring hearing is also warranted, our data suggest that a monitoring program that includes only hearing is insufficient.

Supported by the Cystic Fibrosis Foundation Therapeutics, Inc., grant number N008681-385013.

Collet, K. 1, 4 ; Sauder, A. 3, 4 ; Lapke, J. 2 Purpose: The pediatric CF program at Children's Hospital of Illinois at OSF Saint Francis Medical Center is composed of 46 patients. Our purpose was to improve the nutrition status of these patients through standardized identification and interventions.

Methods: Pediatric nutrition categories as defined by the 2001 CF Pediatric Nutrition Consensus Conference were used, with the addition of an Optimal category (>50th percentile BMI/age). Every patient was assigned to a nutrition category at each clinic visit between Jan 08 and Dec 08. Individual genetic potential was identified. This information was then discussed with each patient/ family. Sixteen of the 46 patients were not included in the data collection: 4 taking prednisone > 6 months, 1 with growth hormone deficiency, 6 with < 12 months data, 4 below genetic potential, and 1 with recent diagnosis of hypothyroidism. A magnetic board divided into the 4 nutrition categories was displayed in the team conference area. Each patient was assigned a magnet color coded to one of three clinic physicians. The magnet was moved to correspond with the assigned nutrition category at each visit. The board was used as a visual prompt to monitor work in progress. A nutritional intervention checklist was developed and used to assess those At Risk or Urgent Need. The checklist identifies the current nutrition regimen and tracked interventions previously tried, on-going, and goals. Interventions included methods to maximize intake, absorption and testing for other medical conditions. Patients and families collaborated with team members on individualized goals and action plans. A follow-up protocol was developed that coincided with the severity of the patient's nutrition status: return in 3 months if Optimal/Adequate; 2 months if At Risk; 1 month if Urgent Need. The CF Center Patient and Family Advisory Board reviewed the process and provided feedback throughout the project.

Results: Growth parameters from 30 patients were tracked throughout the year. Five patients moved in a positive direction, with 1 moving favorably 2 categories. Patients did not experience a decline in nutrition categories. Over the year span the percent of patients in Optimal or Adequate categories increased from 67% to 77%. Throughout the process, many patients and families obtained greater knowledge about their nutrition status. The process also changed the CF Center team's behavior and awareness of the nutrition status of each patient.

Conclusion: Through standardized identification of patients at nutritional risk and standardized interventions, the nutritional outcomes of the pediatric population improved. [1] , ACT in some form is an important component of cystic fibrosis (CF) pulmonary care both chronically and in the hospitalized patient. In the hospital setting, ACT is often performed inconsistently and indifferently and may be given a lower priority than other activities. As part of an improvement project led by a multidisciplinary steering committee (that included parents) intended to make hospital treatment of pulmonary exacerbations more efficient and productive, we developed a program to provide and teach to patients consistent high level ACT.

Method: We began with a team benchmarking visit to Cincinnati Children's Hospital because of the work they had previously reported in this area [2] . We devised a plan to train staff respiratory care practitioners (RCPs) to adhere to a standard protocol for use of Acapella, huff cough, high frequency chest wall oscillation, and autogenic drainage, beginning with the first two. RCPs worked one-on-one with the staff educator to learn a standardized approach to ACT and were asked to demonstrate skills using a standard protocol checklist; they will be re-evaluated yearly. A pilot program was then developed for application to patients 10-19 years old and physically able to perform their chosen ACT as well as huff cough. On admission, patients worked with nursing to set a daily schedule for ACT and other activities, and chose, in collaboration with the RCP, which form of ACT they preferred. They were then evaluated and educated on proper ACT technique. Members of the healthcare team, including physicians, were asked not to disturb the patient when ACT was being performed. Following the correct performance of each treatment per standard protocol, patients were entitled to receive a token. These tokens were exchanged for prizes at discharge.

Results: Three months after initiation of the program, 32/59 (54%) of RTs have been checked off on performance of huff cough, and 25/59 (42%) on use of the Acapella valve. There were 16 patients who were part of the pilot program, and 29 comparable CF patients admitted to the hospital during the first 5 months of 2009 who were not part of the program. As a result of improved scheduling, patients included in the pilot program received 4 ACTs per day (as ordered) on 72% of their hospital days, compared with 59% in the non-pilot patients (p=.09 by ttest). Huff coughs were incorporated into the therapy session 4 times per day on 52% of the days in pilot patients, compared with 18% in the non-pilot patients (p<.001 by t-test).

Conclusion: While these pilot results are derived from a relatively small number of patients over a short period of time, they suggest a fair degree of success in reaching our goal to standardize ACT. We plan to implement the program on all appropriate CF patients and will continue to monitor and refine the program. Supported in part by CFF grant SCHECHQI0. Background: Physical activity promotes airway clearance and increases bone density, strength and endurance as well as overall life expectancy for people with cystic fibrosis (CF) [1] . As part of an improvement project led by a multidisciplinary steering committee (that included parents) whose intent was to standardize hospital treatment of pulmonary exacerbations and make it more efficient and productive, we developed an individualized exercise program for each patient. While previously each CF patient received physical therapy (PT) no more than 3 times/week, under the new program each patient received guided PT 5 times/week and was given an individualized exercise program based on their initial evaluations. A token incentive program was introduced to promote compliance with all treatments included in their schedules. We sought to measure the benefit of this program on patients' physical wellness by comparing aerobic endurance on admission and at the end of seven physical therapy treatments.

Methods: A convenience sample of patients who were admitted for a CF exacerbation was enrolled in the early phase of the project. Each patient received an admission evaluation which included an aerobic endurance test utilizing a treadmill or a recumbent bike. Their maximum heart rate at peak exertion (MHR PE ) was recorded, and the percent of their predicted heart rate maximum [2] was calculated. Upon each patient's seventh guided physical therapy treatment, the patient's MHR PE was again recorded and the percent of predicted calculated. The result of their initial evaluation was compared with the seventh day results in order to assess the efficacy of daily PT on each patient's aerobic endurance.

Results: We report on the results of the first 16 patients evaluated, 11 of whom were part of the intervention group. Eight were female and average age was 15.44 (range 8-21). The mean increase in MHR was 9.16% for participants and 4.68 % for non-participants (p = .51 by t-test). Three patients in the intervention group showed no improvement in their aerobic endurance capacity after seven physical therapy treatments, and 8 did (3 improved by ≤ 5%, 2 improved by 6-10%, and 3 increased by more than 10%). Of those who did not participate during the same time period, 2 did not improve, and 3 did (1 improved by ≤5%, and 2 improved by 6-10%) (p = .51 by Fisher's exact test).

Conclusions: While our current numbers are too small to draw any conclusions with statistical certainty, this pilot data supports the concept that patients who are involved in a comprehensive and proactive inpatient exercise program will demonstrate improvements in their overall aerobic endurance capacity. Based upon these preliminary results, this program will be implemented into the care plan of all of our CF patients admitted to the hospital for treatment of a pulmonary exacerbation. Supported in part by CFF grant SCHECHQI0. year to a distant CF center may be economically and socially prohibitive. In the province of British Columbia (BC), Canada, 69% (82/118) of pediatric CF patients live within 150 kilometers (93 miles) of the CF clinic at BC Children's Hospital (BCCH) in Vancouver; the remainder are spread across a sparsely populated expanse of 944,735 km 2 (340,105 miles 2 ) and may live as far as 2400 km (1500 miles) from the clinic. In 1997, as part of a provincial initiative to offer health services "closer to home," we initiated a yearly CF outreach clinic (CFOC) to Prince George (northern BC, 1710 km or 1062 miles round-trip from Vancouver), adding later a twice yearly CFOC in Kamloops (Interior BC, 708 km/ 440 mi round trip from Vancouver). Funding for staff has been provided by the Canadian CF Foundation (CCFF) and the B.C. Northern Isolation and Travel Assistance Outreach Program. CFOCs are offered in winter months when travel is generally most difficult. Families must travel to the BCCH CF clinic for all other clinic appointments. Our clinic policy in line with CCFF guidelines calls for a minimum of 4 CF clinic appointments per year.

Aim: To evaluate social and economic costs and satisfaction of families attending CFOCs as compared to attending the CF clinic in Vancouver.

Method: We designed a questionnaire which was sent to 20 families who had attended at least one CFOC: 65% responded.

Results: Distances traveled, time spent and costs per family are as follows (all data are for return journeys): distance travelled to BCCH CF Clinic: average: 1,400 km (range 700-2400km); distance travelled to CFOC: average: 190 km (range 8-600km). Average travel time to BCCH: 13 hours (range 7-28 hours); average travel time to CFOC: 2.3 hours (range 0.2-6.5 hours). Average cost to attend BCCH clinic: ca. $490 (range ca. $100-$1200); average cost to attend CFOC: ca. $43 (range ca. $0-$200).

Most families received financial assistance for a portion of travel to the BCCH clinic. The majority (9/13) of families arranged time off work in order to attend clinic, but they were able to take less time off when attending a CFOC. Similarly, while most families were required to find childcare for healthy siblings, they identified a reduced burden in this area when attending a CFOC. All families indicated that a CF physician should attend all outreach clinics; 85% indicated a physiotherapist and 64% indicated a nurse should also attend. Despite families indicating that outreach services greatly reduced their burden, 77% of families wanted their child to continue to be assessed at BCCH in Vancouver at least twice yearly.

Conclusion: As expected, there is a greater cost in social and economic terms to attend BCCH CF clinics. However there still may be costs and travel needed to attend outreach clinics. Despite the vast geographical distance and increased costs, families still want at least twice yearly assessments at BCCH in order for their child to receive what is perceived as a "more thorough" assessment. We will use this data to further evaluate and develop our current outreach services. Background: Patients with cystic fibrosis (CF) have a significant treatment burden. Interventions made in clinic have the potential to improve treatment adherence in patients with CF, but should be targeted to patients' specific needs. Objectives: 1) Evaluate level of adherence in our adult CF population; 2) Discover patient-reported barriers to adherence; 3) Gather patient-reported suggestions for methods to improve adherence.

Methods: This Quality Improvement project was approved by our IRB and included the development, administration, and evaluation of a treatment adherence questionnaire for patients attending our adult CF center. A draft was reviewed by the multi-disciplinary care team and 3 patients. Edits were made based on feedback. It consisted of 2 multiple choice questions for each of 6 treatments (dornase alfa, chest physiotherapy (CPT), azithromycin, nebulized antibiotics, vitamins, and pancreatic enzymes). The multiple choice questions included level of adherence and reasons for nonadherence. Patients responding that a treatment was used "occasionally" or "never" were classified as non-adherent for a given treatment. Free text questions were included to collect patient opinions on barriers to adherence, things that promote adherence, and suggestions for interventions in clinic. The survey was mailed to all patients and responses were anonymous.

Results: A total of 78 surveys were mailed and 27 completed surveys were collected. Among the respondents, non-adherence was highest for CPT (39.1%) dornase alfa (29.2%), and vitamins (26.1%) and lower for nebulized antibiotics (15.8%) and pancreatic enzymes (8.7%). No one reported non-adherence with azithromycin. Perceived barriers to adherence included: time limitations, not having the medication on hand, cost, inconvenience, adverse effects, and lack of an immediate benefit. Things that were reported as helpful included: having a routine, understanding benefits of treatment, accessibility of medication, and assistance of family members. Suggestions for interventions in clinic included: pill dispensers that can fit in a purse or pocket, a discussion of treatment benefits, encouragement, help with time management, and help with cost. We are currently planning to implement interventions in clinic based on these results.

Conclusions: Patients have valuable suggestions for improving adherence. Limitations to our study include responder bias and the fact that these adherence rates are patient-reported. However, the rates of non-adherence for particular treatments in our patient population are similar to those found in other studies in adults with CF. Further studies should be done to measure the impact of simple interventions in clinic on overall treatment adherence. Background: An alternative model of care for physiotherapy contact with children with cystic fibrosis (CF) was piloted to improve on existing access through the multidisciplinary out-patient clinic and in-patient treatment. This entailed provision of a dedicated full-time out-patient respiratory physiotherapy service.

Objectives: To assess the effects of a 12 month out-patient physiotherapy intervention (OPI) in children with CF. The primary outcome measure was hospital admission days with secondary outcome measures of lung function, exercise testing and body mass index (BMI).

Methods: Children who had 4 or more admissions in 2007 (n=10) were selected from the CF clinic population (n=96) for the 12 month OPI pilot that ran in 2008. This was the target group (TG). The TG received regular OPI treatments every 1-3 weeks, combined with acute management plans. OPI treatments aimed to address the patient's individual needs and depended upon the patient's age and condition but consisted of the following: airway clearance, nebulised mucolytic treatments, aerobic exercise, strength training, stretching and postural advice, and an individualised home exercise and airway clearance program. Regular spirometry (SpiroAir, Medisoft, Belgium) and an incremental 20 meter shuttle test were performed by all eligible patients.

Results: Ten children (7 female) formed the TG and ranged in age from 3-18 (mean 13.2) years. Seventy percent were homozygous for the F508del mutation with the remainder compound heterozygotes. Seven of 10 were able to perform spirometry and 4/10 the shuttle test.

Data for 2007 were compared with those from 2008 for admission days, BMI as well as lung function parameters. These are displayed below (Table  1) .

Discussion: A group of children with CF targeted with intensive outpatient physiotherapy over a 12 month period exhibited reduced hospital admissions, as well as improved lung volumes and shuttle test scores. The TG had a total of 349 admission days in 2008 compared with 677 in the preceding 12 months.

It appears that intensive out-patient physiotherapy may exert a beneficial effect on reducing hospitalisation in CF with likely cost-benefit effects. Additionally a benefit on halting rate of decline of lung function and improving exercise ability is suggested. Further research is planned to prospectively evaluate the OPI scheme, with detailed cost-benefit analyses, and a wider range of outcome measures including exercise capacity and quality of life. Airway clearance, an essential element for maintaining and improving the lung health of people with cystic fibrosis (CF), is a mainstay of pulmonary care. To improve the lung health of our patients through the use of age-appropriate airway clearance techniques, our CF Center embarked on a quality improvement (QI) initiative. The specific aim was to increase the number of patients receiving formal airway clearance education. In particular, our plan was to educate 25% of all patients seen in the CF Center from 4-1-08 to 7-1-08 and 100% of all patients seen from 4-1-08 to 3-31-09. To that end, a multidisciplinary team, including a parent representative, was convened and developed an airway clearance training protocol, a patient incentive plan, patient educational materials and an airway clearance questionnaire. Additionally to assure that the entire CF team provided uniform education and a consistent message, three airway clearance training sessions were scheduled for the entire CF team. Checklists for each form of airway clearance were developed and an emphasis was placed on the importance of daily exercise. Baseline adherence, obstacles to therapy, lung function measures and BMI were performed prior to instituting the airway clearance project and will be reevaluated at the end of the study period. From 4/1/07 to 3/31/08, only 15% of pediatric patients had documentation of having received airway clearance teaching (ACT). In contrast, a year after instituting the project, 89% of patients had documentation of having received ACT. The most common form of airway clearance utilized by our patients was high frequency chest wall oscillation (HFCWO), with 33 patients (51%) using HFCWO exclusively. A significant number of patients report that they utilize multiple airway clearance modalities (n=19, 29%). Lastly, a small number of patients report using the Acapella (n=6, 9%), the manual percussor (n=3, 5%), or CPT (n=1, 2%) exclusively. Three (5%) patients did not respond to this question. With regard to adherence twelve (18%) individuals report not having enough time during the day between work and school as an obstacle to airway clearance. An additional four patients (6%) did not feel airway clearance worked or did not think they needed it when they felt well. There were 3 subjects (5%) who found airway clearance embarrassing. Lastly, 21 respondents (32%) listed "other" reasons for not performing airway clearance as prescribed. The remaining study parameters including FEV1 and BMI, will be evaluated at the 12 month mark for each patient. The interim results from this project indicate that a standardized team approach to airway clearance training can increase the number of patients who receive airway clearance training. Lessons learned from this QI initiative include the facts that pre-planning is necessary, teamwork is important, communication is vital, and family involvement is critical. Armoni, S.; Rony, R.Y.; Cohen-Cymberknoh, M.; Shoseyov, D.; Pugatsch, T.; Kerem, E. CF Center, Hadassah Medical Center Mount Scopus, Jerusalem, Israel Background: Patients with cystic fibrosis (CF) need attention, financial resources and energy for their daily, time-consuming therapies to maintain an optimal health status. Our goal as healthcare providers (HP) is to offer the best care. This results in a heavy burden for the patients which impairs the quality of life (QoL) and may lead to poor adherence. The ultimate goal should be to propose a true balance between quality of care (QoC) and QoL. The QoC in CF is measured by outcome variables that include morbidity, mortality, FEV 1 and BMI. Over the last decade, the patient's perspective on QoL was included as well. Patients grade the effect of the disease on their daily life, and include their subjective assessment of respiratory status, physical activity, eating, body image, social status and the emotional effects of CF.

Aim: To investigate the differences in perception of quality of life and quality of care between HP and patients/parents in our CF center.

Method: QoL and QoC were assessed by an anonymous questionnaire that included 46 questions, completed by 13 HP, 24 patients >18 years and 29 parents of children <18 years.

Results: No significant differences between the perceptions of HP and the patients on QoC were detected. Significant differences were observed in QoL, i.e., subjective perception, daily activities, treatment burden and coping.

Despite the original assumption that patients perceived their QoL much lower than expected by the HP, this study showed the opposite.

Conclusions: Patients and parents perceive the chronic illness in a "healthier" light than the medical staff. Therefore healthcare providers should take the patient's perception into consideration and recommend the best care when developing treatment regimens. Based on these results the research will be expanded to include additional CF Centers in Israel. Background: Since 1998, high frequency chest wall oscillation (HFCWO) therapy has been an integral part of the treatment program at our Cystic Fibrosis Center's pediatric and adult programs, often beginning at 2 years of age. Our usual recommendation is that HFCWO be used once or twice daily for 20-30 minutes each time. Our instructions are to increase the oscillation frequency in 1 Hz increments from 10 to 14 Hz with a "huff" cough between each frequency change, and at the end of the session. Patients are told that a "huff" cough involves a maneuver similar to blowing one's breath in order to cause "steam" to appear on a mirror. At each clinic visit, patients or their parents are asked to report the hours of HFCWO as recorded on the machine. Documented adherence is rewarded with a $10 gift card. Once a year, our respiratory therapists review all respiratory treatments and modalities with patients and/or parents. In order to assess the efficacy of our educational efforts relating to the rate of adherence to HFCWO therapy, a survey was administered by our respiratory therapists.

Methods: A convenience sample of 78 patients and/or parents (42% of our Center population, ages 4-57) were asked the following questions regarding HFCWO: Is its frequency varied sequentially or at all? Is a "huff" or any kind of cough initiated between frequency changes or at the end of the therapy?

Results: Only 10% of respondents varied the frequency and intentionally coughed between frequencies or at the end of the therapy; 49% varied the frequency but did not cough; 6% intentionally coughed at the end of the therapy but did not vary the frequency; and 35% neither varied the frequency nor coughed.

Our stated goals for HFCWO therapy have not been met despite significant efforts. We speculate that the reasons for this short-coming include that when HFCWO is initiated at 2 years of age, it cannot be expected for toddlers to cough voluntarily. When these patients became older, it appears that we may not have conveyed adequately the importance of coughing, or that the parents and patients had difficulty changing their physiotherapy routine. Based on the survey results, we have begun to explain to our patients and their parents that without coughing, HFCWO is analogous to sweeping the floor without picking up the debris, which allows it to rescatter on the floor. After hearing this analogy, a number of our patients have reported that their HFCWO now helps them to have a productive cough. Further, we are creating laminated cards detailing the desired procedure and importance of HFCWO in addition to other respiratory procedural techniques. We plan to distribute this information to all of our patients, and to review it with them at each quarterly visit to our Center. The CF MT focuses on high impact therapies: Pulmozyme ® , Tobi™, Hypertonic Saline, and Azithromycin. An additional area of focus of the MT was to evaluate the use of leukotriene receptor antagonists and inhaled corticosteroids. Minimum criteria for anti-inflammatory use was 12% reversibility in FEV1 or clinical wheezing in the prior year. Ninety-one unique patients between 6 and 16 years of age had at least one MT encounter in each of the years: 2006, 2007, and 2008 and were evaluated. The 2006 values were considered baseline values. Lung function and nutritional parameters were compared between the baseline MT and 2007 and 2008. Median FEV1 per year was calculated using the median of the mean of the best quarterly values. BMI percentile was calculated as the median of the mean of best quarterly BMI%.

Results: Fifty-one of 91 patients were eligible for Tobi™ with 100 %prescribed. Seventy-three of 91 patients were eligible and prescribed Pulmozyme ® . Twelve of 91 patients were eligible and not prescribed; 11/12 were prescribed hypertonic saline as an alternative. Thirty-one of 91 patients were eligible for azithromycin and prescribed. Fourteen of 91 patients additional patients were eligible and not prescribed and 9 varied in eligibility during the 2 years. Seventy-five of 91 patients were eligible and prescribed hypertonic saline during data collection. Fifteen eligible patients were never prescribed hypertonic saline. Eighty of 91 patients met our minimal eligibility for anti-inflammatory therapy. Seventeen of 91 patients were prescribed montelukast and 54/91 were prescribed inhaled corticosteroids +/-montelukast. FEV1 decreased from 99.9 (2006) to 98.8 (2007) to 97. 1 (2008) . Median BMI varied from 43% to 44% to 41%.

Conclusions: During this period of data collection there was very high compliance with prescribing of Tobi™ and Pulmozyme ® . There was also high compliance but slightly more variability with azithromycin, hypertonic saline, and anti-inflammatory therapy. Defining eligibility for therapies can become challenging and tends to vary over time. Despite high compliance with medications as demonstrated with the MT, lung function and nutritional status declined slightly in this population over time. These data do not reflect the individual changes in clinical care made directly as a result of the MT. We demonstrated a high level of compliance with CF medication prescription could be achieved. The MT often triggers a change in medication regimens to assure compliance. We speculate a modification of the tracker with increased emphasis on microbiology and lung function trending will result in better preservation of clinical status over time.

Duval, M.A.; Patel, K.; Eaton, Z.; Pilzer, E.; Crews, B.; Scott, P.H. Children's Healthcare of Atlanta, Atlanta, GA, USA Background: The rate of lung function decline in cystic fibrosis [CF] patients is correlated strongly with nutritional status. Maintaining weight gain in patients with CF is a challenge, in part, because of a decrease in appetite. The appetite stimulant, megestrol acetate (megestrol), has been used in CF patients who exhibit decreased appetite and poor weight gain. As part of a quality improvement initiative -developing an outpatient nutrition pathway -we performed a retrospective evaluation of the use of megestrol in our care center.

Methods: We identified all patients treated with megestrol from January 2004, to December 2007. We evaluated dose, duration of therapy, weight, height, body mass index [BMI], lung function, blood glucose level and cortisol level.

Results: Twenty patients (20 months to 17 years of age) received megestrol during the study period and were included in the analysis. The average initial dose was 5mg/kg/day (1.5 -10mg/kg/day), and the average final dose was 4mg/kg/day (1 -9mg/kg/day). The mean duration of therapy was 10.4 months (1.5 -36 months) . Seventeen patients were on therapy for 12 months or less, and 3 patients received megestrol for more than 12 months. The average weight gain (+/-standard deviation) was 4.9 +/-4.2kg (0 -18kg), with an average percent weight gain of 20% (0 -37%). The average height gain was 5.6 +/-5.9cm (0 -22.3cm), with an average percent height gain of 4.8% (0 -15%). The average increase in BMI percentile was 29.6 +/-22.2 (1 -74) . For the 12 patients old enough to perform lung function tests, the average pre-megestrol forced vital capacity [FVC] was 86.9 +/-16.2 percent predicted [% pred] (55 -108% pred), and the average forced expiratory volume in one second [FEV1] was 74.9 +/-20.4% pred (40 -104% pred). Lung function values at the end of therapy included an average FVC of 92.9 +/-18.6% pred (55 -121% pred), and an average FEV1 of 80.3 +/-21.1% pred (53 -121% pred). We observed an increase in average FVC of 6.0% pred and an increase in average FEV1 of 5.4% pred. No patients had cortisol levels checked during the study period. Fifteen patients (75%) had a blood glucose level checked. Seven patients had elevated blood glucose levels, including 2 patients who had a diagnosis of cystic fibrosis-related diabetes [CFRD] before starting megestrol. One patient was diagnosed with CFRD during the study period.

Conclusion: Our data show a variation in megestrol dose and duration of therapy and in monitoring for adverse effects. Overall, the use of megestrol was associated with an improvement in nutritional status and lung function. To standardize the use of megestrol, our care center has developed an appetite stimulant protocol, which is incorporated in a new outpatient nutrition pathway. Per the protocol, the starting dose of megestrol is 5mg/kg/day. A weight goal for each patient is established, and if the weight goal is met, megestrol is discontinued. If megestrol is continued, a morning cortisol level and blood glucose level are checked every 6 months. The initial focus of our quality improvement program was to improve the nutritional status of our patients with cystic fibrosis (CF). Initially we placed each patient into a nutritional category, which is now tracked and adjusted at each visit, then assuring that every patient left clinic with a plan laid out for nutrition until the next visit. The next step was to standardize our approach and assure that we are not missing any important issues for those that are not at goal. We developed a nutrition action plan and as a team worked our way through a preset plan for all those not at goal nutritional status. We have tracked results both for our interventions and for BMI of all our patients. Categorization is reliably done for 95-100% of all patients at each visit, and, despite a large workload associated with every patient at every visit, we have placed written nutritional action plans in the hands of approximately 95% of all patients seen, at each visit. We feel that these numbers represent excellent compliance by our staff. We have not yet begun tracking use of the intervention checklist, but feel that this will be the hardest to both implement and track.

Surprisingly, we have already seen a shift upwards through the nutritional zones, and a shift of approximately 6% of our patients out of the zone of actual nutritional failure, into better status categories. That represents a movement out of nutritional failure into a category of nutritional risk for approximately 20% of those most poorly nourished. We are very pleased by this. In addition, within the larger category of nutritional risk we have seen a shift upward, more toward satisfactory nutrition. We are utilizing a grant from the CF Foundation to establish an incentive program for patients, hoping to further increase weight gains.

We have also begun the process of assessing and intervening in areas that we have identified as important for improved pulmonary function: compliance with appropriate and correct airway clearance therapy and compliance with proper equipment cleaning. We are looking at improvement in these areas both in the inpatient and outpatient settings. We have now reinstructed 44% of all outpatients in these areas, and have elaborated a process to assure improvement when patients are inpatients, as well. The next step will be tracking the degree of success in making change. We are also auditing outpatient charts to determine if there is sufficient variance among providers in treating pulmonary exacerbations to warrant a more systematic approach to treatment of worsened symptoms, in the clinic.

The Patient Family Advisory Board (PFAB) has met monthly this year and has established Bylaws. They are now assessing what role they want to play, and are receiving formal "Volunteer" training. There are now several family members, and one adult patient on the board, complemented by a CF Center social worker, with guest staff members, as needed. We also have a parent on our Quality Improvement team. He also attends the PFAB.

Our plans for the next quarter are to proceed with these activities, with possibly revising the way we do clinic review. In 2001, the Cystic Fibrosis Foundation made specific recommendations for infection control practices for patients with cystic fibrosis (CF). In response to these recommendations, our center engaged in a quality improvement initiative to optimize our infection control and prevention practices with the following global aim: We aim to improve infection control and prevention in the lives of our patients with CF. By working on this process, we expect to reduce the acquisition and spread of respiratory pathogens. It is important to work on this now to minimize the risk of acquiring new or resistant respiratory pathogens.

In order to achieve our aim, a multidisciplinary QI team, including family and patient advisors, convened using a microsystems methodology to approach infection control and prevention. A comprehensive process for infection control was developed, including but not limited to patients wearing masks and using hand gel, cleaning of the exam rooms and equipment between patient use and the use of gowns and gloves by all providers for all patients. Education specific to our process was performed with all staff, patients and families. The process was implemented with the pediatric population in October 2007. Our specific outcome measure was the percent of positive cultures per organism by quarter using data extraction from Port CF.

Prior to the onset of this initiative our bacterial culture rates had remained steady over the past 5 years and reflected the average baseline rate of bacteria found in CF cultures nationally. Since the implementation of our new infection control processes, pediatric cultures at our center have had a 10% decline in cultures positive for Pseudomonas aeruginosa , and a small but steady decrease in % cultures positive for methicillin-resistant Staphylococcus aureus from consistently above 10% to being consistently less than 10% of all cultures. Other bacterial rates at the pediatric program have had small declines to become consistently lower than national rates. The adult program has subsequently implemented infection control practices and cultures rates have been stable to date, with rates of Stenotrophomonas maltophilia, MRSA and Burkholderia cepacia below the national average.

The implementation of more stringent and consistent infection control processes has been successful in reducing the acquisition and spread of respiratory pathogens in the pediatric population, and has been well accepted and welcomed by patients and families. These are sustainable processes in both inpatient and outpatient settings and have become our new standard of care. Continued monitoring of pathogen acquisition and culture rates, as well as periodic assessments of adherence to infection control processes will ensure that our aim to prevent the spread of CF pathogens and ultimately preserve the lung health of our patients will be met.

Gunn, E. 1 ; Osmond, J. 1 ; Baker, M. 1 ; Larsen, A. 1 ; Bilton, D. 2 1. Cystic Fibrosis Trust, Bromley, United Kingdom; 2. Royal Brompton Hospital, London, United Kingdom The Department of Health is investigating the possibility of a funding system based on a whole year of care for each patient with cystic fibrosis (CF) in England, depending on disease severity and treatment requirement. We have developed a model consisting of seven costing bands. Band 1 reflects mild disease with no requirement for nebulised therapy. Band 1A is triggered by requirement for Pseudomonas eradication therapy. Patients in Band 2 and 2A receive DNase or long term nebulised antibiotics (Band 2) or both (Band 2A). The triggers for higher bands (Band 3-5) include total number of days on IVs, total days spent in hospital, gastrostomy or nasogastric feeding, CF-related diabetes, pneumothorax, massive haemoptysis and ABPA.

The introduction of Port CF software to the UK CF Registry has facilitated the calculation of a band for each individual patient based on the annual review. The aims of this pilot study were to assess the distribution of bands in both paediatric and adult clinic populations and to assess the change in bands over 2 consecutive years. A parallel costing exercise is ongoing.

Methods: Thirteen CF Centres were recruited for the costing and banding exercise. Each patient was assigned to a band from the annual survey data for the years 2007 and 2008. Costs of care were also calculated for each patient.

Results: There were 884 adult and 528 paediatric patients with 2 years of consecutive data from the 13 centres (adult=6, paediatric=7). The distribution of bands for 2007 and 2008 are shown in Table 1 . A total of 269 (31%) adult patients demonstrated a change in band from 2007 to 2008. In the adult population 180 patients showed a change to a higher band, signifying increased requirement for therapy. In contrast 89 patients demonstrated change to a lower band in consecutive years signifying a reduction in intensity of treatment. In the paediatric population 219 (42%) demonstrated a change in band with 122 changing to a higher band signifying increased requirement for therapy but 65 having a reduction in band. Change to a higher band was not necessarily associated with significant decline in lung function, reflecting the approach of increased intensity of treatment designed to prevent loss of lung function.

Conclusion: This exercise demonstrates that each patient's band can be derived objectively and consistently from clinical data held in Port CF. This data can be used to derive a funding system for CF centres once average costs per band have been established.

Guasco, G. 1,2 ; Stokes, D.C. 3, 4 ; Joyner, R.E. 2 ; Griffin, L. 2 ; Hunsicker, J. 2 ; Bowdre, A. 2 ; Nason, L. 2 ; Warren, L. 5 ; Burness, G. 5 1. Quality Improvement, LeBonheur Children's Medical Center, Memphis, TN, USA; 2. Laboratory, LeBonheur Children's Medical Center, Memphis, TN, USA; 3. UT CF Care and Research Center, LeBonheur Children's Medical Center, Memphis, TN, USA; 4. Department of Pediatrics, UT Health Science Center, Memphis, TN, USA; 5. Nursing, LeBonheur Children's Medical Center, Memphis, TN, USA The aim of this project was to improve our sweat collection technique so that we achieve our goal of less than a 5% "quantity not sufficient" (QNS) rate, a requirement for maintaining accreditation with the Cystic Fibrosis Foundation and the College of American Pathologists. A multidisciplinary improvement team focused on recognizing issues that could potentiate the collection of an insufficient sweat sample during the preparation, stimulation, and collection phases of pilocarpine iontophoresis. The project scope covered all ages undergoing sweat chloride testing but did not include the analysis phase. To decrease insufficient samples, the team utilized Six Sigma/LEAN process tools to evaluate and control the collection process. The DMAIC process, which includes the following steps, was used: Define (purpose of project), Measure (collect and summarize data), Analyze (determine the causes of variation), Improve (determine best solution to minimize variation and reduce defects in the process), and Control (assure improvements are sustained). After defining project goals and collecting baseline data, process flow and fishbone diagrams helped delineate variations in technique during the sweat collection process. A detailed analysis of each step of the process was reviewed, evaluated for variation, and standardized. Opportunities for improvement were identified, including a change in filter paper size and a different method of sealing the stimulated area to minimize evaporation. Staff education was performed, and awareness of insufficient sweat samples was raised through posting compliance, one-on-one meetings with staff to provide feed back concerning individual performance, and celebrations for the staff when milestones were achieved. Education was developed for parents with the assistance of the Patient and Parent Advisory Council regarding patient preparation prior to collection. To sustain these changes, a control plan was implemented to assure continued success. The overall QNS rate fell from 6.92 per 100 patients in 2008 to 2.47 after the test of change was initiated, a 64% reduction. Ultimately, we expect to see a statistically significant improvement as our sample size increases. registry data, our median BMI was less than the national average in our pediatric population. Therefore, we implemented several specific change ideas between April and June of 2008: 1) developed a written communication tool for patients and families about their nutritional and pulmonary plan of care as well as written goals for weight and pulmonary function; 2) created an Electronic Medical Record (EMR)-based nutrition, pulmonary and therapy summary tool for pre-clinic screening; the EMR screen includes BMI percentile, FEV1, sputum cultures, diabetes and DEXA screening dates, as well as pulmonary therapies (DNase, hypertonic saline, azithromycin, inhaled antibiotics); 3) implemented a new process to ensure 100% of patients complete yearly diabetes screening beginning at age 13 by obtaining 2-hour post Glucola glucose levels during annual clinic visits; 4) established a process for the dietitian to prioritize contact time for patients with BMI < 50th percentile; 5) developed and distributed a questionnaire to assess adolescent CF care knowledge, permitting us to provide targeted education in a monthly STAR (Starting Transition to Adult Responsibility) adolescent clinic. We assess our patient and family involvement at a level 3 during this quality improvement (QI) process.

Results: We have followed the percent of patients meeting BMI (≥ 50th percentile) and FEV1 (≥ 98% predicted) guidelines. The percentage of patients ages 6 through 19 (N=61) meeting CFF guidelines for BMI increased from 28% (Q1) to 48% (Q4) (+20%) throughout 2008, a statistically significant change (p-value=0.01). Percent of patients meeting CFF guidelines for FEV1 also increased from 30% (Q1) to 41% (Q4) (+11%) by the end of the year but this change is not yet significant (p-value=0.26).

Conclusion: Participation in LLC VI produced measurable improvements in nutritional outcomes in our care center, with a trend towards improving lung function. Maintaining enthusiasm for improvement after the end of the collaborative has required continued regular QI meetings, sharing improvement data with patients and families at education day and patient and parent advisory board meetings, and incorporation of our aims into routine clinic practice. Introduction: Certain behaviors associated with feeding and mealtimes can become a barrier to achieving adequate energy intake and optimal nutri-tion for young children with cystic fibrosis (CF). Studies have demonstrated that nutrition counseling and interventions are most effective when provided at a younger age. The Behavioral Pediatrics Feeding Assessment Scale (BPFAS) has been used to identify behaviors associated with inadequate nutritional intake (1) . At our center, a nutrition recovery protocol (NRP) was created for patients who have suboptimal nutrition status (SNS), defined as wt/lg% or BMI% <50. A specialized clinic, staffed by a nurse practitioner, RN, RD, and licensed counselor (LC), was developed for patients with SNS and/or behavioral issues affecting nutrition as identified by the care team or parent. This Nutrition Clinic (NC) was specifically designed to address behavioral issues and co-morbidities, such as sinus disease or medication interactions, that impact nutrition. The purpose of NC was to improve the nutrition status of patients with SNS by providing focused clinical nutrition interventions and incorporating a counselor to address parent and child behaviors as a unique intervention.

Purpose: We aimed to characterize the population referred to the NC and describe outcomes.

Methods: In 2006, patients with SNS ages 14 mo-14 yrs were identified and referred by their CF physician to the NC. With the NRP as a guide, patients were scheduled to attend six NC visits. The BPFAS was used to identify behaviors associated with poor nutritional intake. The LC provided behavioral counseling for each patient. Examples of NC interventions included weekly introduction of new foods, consistency of meal/snack schedule, and avoidance of preparing a special meal only for the patient. In addition to routine assessment, diet records were analyzed at each NC visit.

Results: Eleven patients (7 male) enrolled in the nutrition clinic from 2006-2008, ages 1-14 yrs. At the time of referral to the NC, wt/age%, ht/age% and wt/lg% or BMI% ranged 0-64%, 0-64% and 3-57% respectively. The patients attended an average of 2.6 NC visits before returning to usual CF clinic, which resulted from improved nutrition status, resolution of eating behaviors, parental inconsistency with behavioral interventions or change in clinical status. Barriers to success were co-morbidities and parental inconsistency. After an average follow-up period of 14 months, wt/age%, ht/age% and BMI% ranged 1-71%, 1-76% and 4-71% respectively. Median BMI% showed significant improvement from 22% to 41%.

Discussion: Consistent with the literature, better outcomes were found in younger patients. Unexpectedly, fewer visits were required to achieve improved outcomes, likely due to the intense focus on nutrition not found in usual CF clinic. An additional benefit to the NC was the counseling available to families who would otherwise have no access to these services. More aggressive recruitment to the NC is planned for patients with consistent SNS. Background: Many factors contribute to poor weight gain in cystic fibrosis (CF). The presence of poorly controlled CF-related diabetes (CFRD) is often associated with weight loss or difficulty gaining weight. Lack of education and inadequate monitoring increase the risk of poor glycemic control. A quality improvement (QI) initiative targeting individuals with CFRD was developed at our centre in an effort to improve the nutritional status of our patients.

Purpose: To ensure adequate monitoring of CFRD patients and to increase diabetic education in our patient population.

Methods: The first stage of this QI initiative was to evaluate reasons for poor weight gain in our patient population. With input from the patient/family advisory board, a questionnaire was developed to identify barriers to weight gain. Reasons for poor weight gain were broad so our team chose to focus on diabetic management as a large proportion of our CFRD patients had suboptimal nutritional status. The second stage of this project involved developing a diabetic education program and assessing the frequency of diabetic monitoring using hemoglobin A1C. A diabetic education checklist was created. All CFRD patients attending the outpatient clin-ic were identified and diabetic education was given by the CF nurse practitioner/dietitian using the checklist. The proportion of patients who received diabetes education at each clinic was tracked. Standard of care at our centre involves monitoring A1C in patients with CFRD every 3-4 months. We began tracking A1C measurements and reasons for inadequate monitoring of A1C were recorded. Strategies were developed in order to improve the frequency of monitoring.

Results: A total of 101/400 (25%) CFRD patients are followed at the Adult CF clinic at St. Michael's Hospital. Forty percent of CFRD subjects were post-transplant. The diabetic education checklist was successfully reviewed with > 95% of CFRD patients who attended the outpatient clinic. If the patient was ill or being admitted to the hospital, diabetic education was deferred until on the ward. Reasons for inadequate monitoring of A1C included patients lost to follow-up, northern clinic patients, and non-compliance. During the study period, pre-transplant patients with A1C measurements within 4 months increased from 45% to 59% while post-transplant patients improved from 18% to 33%. Since starting the QI initiative, the mean A1C for non-transplant patients has gone from 7.6% to 7.2%. Data on changes in BMI of our patients with CFRD is currently being collected.

Conclusions: We developed a diabetic education tool and improved the monitoring of our CFRD patients as part of a QI initiative. These efforts have increased diabetic education and monitoring. A1C has decreased slightly and in time, we anticipate that this will translate into improved nutritional status for our CFRD patients. Post-transplant patients and those lost to follow-up are at highest risk for inadequate monitoring. Strategies to target these two groups are being developed because of this QI initiative. Optimal nutritional status is a target quality improvement goal for our pediatric cystic fibrosis (CF) center, serving approximately 225 infants, children, and adolescents. The nutritional status of children with CF hinges on the ability of parents to effectively manage dietary issues associated with CF. Previous studies at this center determined that parents' knowledge and confidence in nutrition-related skills were lower than in other components of CF care. We hypothesized that empowerment of parents with CF nutrition knowledge results in improved nutrition status for their children. This participative research project aimed to identify parents' perceived needs for CF nutrition information and their preferences for receiving the needed information. The local CF Parent Advisory Board (CFPAB) collaborated with this research project through feedback at CFPAB meetings and administration of weekly email correspondence with parents.

Three survey methods were used: 1) a preliminary online survey, sent to parents who received weekly "CF Nutrition Tip of the Week" emails (n=12), 2) a written survey, developed from the online survey and distributed at a CF parents' educational event (n=33) and 3) a focus group of interested parents who were not CFPAB members (n=8).

The online survey elaborated on parent responses to a quality improvement questionnaire conducted on the CFPAB website. The online survey was designed to obtain more in-depth information regarding nutrition-related topics of interest and electronic formats preferred by parents.

The written survey developed themes from the online survey. In the written survey, parents received a list of 12 CF-related nutrition topics and rated their interest (0-4) in each topic. In order of parents' preference, the topics were: 1) CF-specific recipes, 2) cooking tips, 3) healthy habits for the whole family, 4/5 tied) meals on a budget and how to get your child to eat, 6/7 tied) mealtime behaviors and exercise/physical activity, 8) adding extra calories to foods, 9) fluids, 10) salt intake, 11) vitamins, and 12) enzyme management. Additionally, parents were asked, "If interested in this topic, how would you like to receive this information?" The format options included: website info, email, Facebook, USPS mail, handouts in clinic, lectures, and phone mail. Across all topics, most parents liked more than a single format. Generally, email was the parents' favored format with 45-65% approval, website 33-39%, handouts 27-33%, USPS mail 12-18%, and lec-tures 12-15%. Facebook and phone mail were selected by less than 10% of parents as preferred formats for delivery of nutrition-related information.

The focus group discussion centered on the preferred length and depth of nutrition information requested by parents. These parents valued frequent, brief information with concrete, age-specific nutrition suggestions they could implement. Although parents did not routinely want in-depth presentations of background, research findings and rationale, they felt this information was important and should be available to parents upon request.

A CF Foundation quality improvement grant funded this research. Nutrition care for the cystic fibrosis (CF) population is a complex and dynamic field. Some CF centers have dietitians that are new practitioners and/or new to CF who may not realize the extensive experience in the CF community and the wealth of available nutrition resources that are openly shared within the CF community. We developed a mentoring program to provide hands-on training, educational resources and strengthen networking within the community for dietitians new to CF with the ultimate goal of improving the quality of nutritional care provided to individuals with CF. The program is supported by the Cystic Fibrosis Foundation.

Method: Two facilitators manage the program. Facilitators regularly update a Nutrition 101 document, match applicants (apprentices) to a mentor (1-2 cycles/yr), and monitor progress of the matches. After a match is made, the mentor contacts the apprentice to discuss clinic characteristics, potential questions, set objectives and assist with scheduling a visit to the mentor's care center. Educational objectives are determined before the visit to target areas to discuss. After the visit, the apprentice, in conjunction with his/her center director, determines a nutrition-related goal with measurable outcome to apply information learned during the interaction to clinical practice. A post-visit evaluation is completed by the mentor, apprentice, and the apprentice's center director or nurse coordinator.

Results: Since the program was launched in the summer of 2007, 52 apprentices have been matched with a mentor and 42 have completed the program. Twenty-one mentors participate in the program (mean of 2 apprentices per mentor, range 1 to 4 apprentices/mentor).

Post-visit summaries from apprentices and mentors have been overwhelmingly positive. One hundred percent of the mentors and apprentices agreed that the program is useful and 98% of apprentices met their educational goals developed for the program.

Center directors and nurse coordinators who responded to post-visit surveys indicated that 90% either strongly agree or agree that the program met expectations. Another 70% strongly agree or agree that nutritional care improved at their center, and 80% strongly agree or agree that the RD was more knowledgeable and active in CF care. Sixty percent of centers who had a dietitian participate in the program developed a measurable goal related to improving nutrition outcomes.

Conclusion: The CF nutrition mentoring program has successfully trained apprentice dietitians and improved delivery of care at their centers. The Nutrition 101 document developed to serve as a foundation for the mentoring program has proven to be an important resource for the entire CF community. The program has also enriched the professional development of the mentors. The mentoring program model has been adapted by several other disciplines in the CF community including respiratory therapists, social workers, and research coordinators with early success.

Background and Aims: Liprotamase (formerly ALTU-135) is a novel non-porcine pancreatic enzyme replacement therapy (PERT) containing three highly purified microbially-derived enzymes: crystalline protease, amylase, and crystalline, cross-linked, recombinant lipase. The efficacy and safety of this PERT has been demonstrated in two well-controlled shortterm studies. We performed a one year open-label trial to study the longterm safety and tolerability of liprotamase in cystic fibrosis (CF) subjects. This trial also measured long-term outcomes such as weight, height, BMI and fat soluble vitamin levels.

Methods: We planned a year-long, open-label study in subjects with CF, some of whom participated in a Phase III efficacy study of this PERT and others who entered the trial de novo. Subjects were ≥ 7 years of age, met cri-teria for CF, were clinically stable, and had a fecal elastase < 100 µg/g. Subjects took one capsule of liprotamase (containing 32,500 units lipase activity, 25,000 units protease activity and 3,750 units amylase activity) in the middle of 3 meals and 2 snacks each day. Subjects were allowed to increase to 2 capsules at the discretion of the PI. Safety measures included frequency and severity of adverse events and laboratory parameters.

Results: After obtaining informed consent, 248 subjects were screened with 217 subjects (58% male) enrolled at 45 study sites (34 in the U.S.; 11 outside the U.S). Eighty-nine subjects participated in the Phase III efficacy study and "rolled over" into this protocol. One SAE was judged by the PI to be reportable to regulatory authorities; this event occurred after an episode of abdominal trauma on day 266 of treatment. The last subject completed the last visit on May 7, 2009 . Database lock is anticipated for June 2009.

Conclusion: Complete results of the safety analysis will be reported at the conference. This study represents the largest prospective trial to date to assess the safety, tolerability and long-term outcomes of a pancreatic enzyme replacement therapy.

This study was supported by the Cystic Fibrosis Foundation Therapeutics, Inc. and Alnara Pharmaceuticals, Inc. Introduction: Depletion of whole body sodium (Na) has been associated with poor growth in animal studies as well as in infants. Infants with cystic fibrosis (CF) have excessive Na and chloride loss through their sweat. Poor growth in infants with CF is associated with poor lung function later in life. Infants with CF are given Na supplements to prevent electrolyte disturbances and poor growth. However, the assessment of adequate supplementation is difficult, as Na depletion occurs before hyponatremia is apparent. Fractional excretion of sodium (FENa) has been used to measure sodium sufficiency, but it requires blood collection. Urinary Na:creatinine (UNa:Cr) ratio only requires a spot urine collection and correlates well with FENa (estimated normal range of FENa 0.5-1.5 corresponds to UNa:Cr 17-52). This study sought to obtain values of UNa:Cr ratio in healthy infants, to study their sodium intake, and to create reference ranges to guide sodium supplementation for infants with CF.

Methods: This cross-sectional study was conducted at a private medical office during scheduled well-baby visits. Inclusion criteria were healthy infants ages 0-12 months with parents willing to provide informed consent (IC). Exclusion criteria were infants with chronic medical conditions (including renal disease and CF) or recent illness with fever, vomiting or diarrhea within two weeks of the visit. After IC was obtained, demographic information, a 24 hour dietary intake survey, and a bagged urine sample were collected. Based on the estimated normal UNa:Cr range (17-52) in a previous study, a sample size of 100 was estimated to produce a 95% confidence interval equal to the sample mean ±1.4 when the estimated standard deviation is 7.

Results: Of 137 patients approached, 121 met criteria and provided IC. Sixty-two patients subsequently were excluded because infants did not produce urine during the office visit. Subjects were divided into dietary (human milk vs. formula) and age groups (0-3, 4-6, 7-9, and 9-12 months) for analysis. Fifty-nine percent (61/104) of infants were fed human milk. The mean Na intake was 182.9 mEq/day (SEM=7.8) for human milk-fed infants compared with 221.4 mEq/day (SEM=12.2) for formula-fed infants. Although mean sodium intake increased with age in the first year of life from 142.7 in the 0-3mo group to 251.4 in the 9-12mo group, the mean UNa:Cr ratio did not show a significant change between age groups. Of the infants with Na:creat ratio <17 (n=39), the mean Na intake was 185.1 (SEM of 7.9) vs a mean sodium intake of 171.1 (SEM 15.6) in those with a ratio > 17(n=9). There was no clinically or statistically significant difference in mean Na intake between these groups.

Conclusions: In this small sample, UNa:Cr is low in healthy infants during the first year of life and is unrelated to Na intake. Data from healthy infants is needed when evaluating measures of Na sufficiency in infants at risk for Na depletion, such as those with CF. Introduction: Cystic fibrosis transmembrane conductance regulator protein (CFTR) controls approximately 80% of gastrointestinal (GI) tract pH. Inadequate proximal small bowel gastric acid neutralization may interfere with pancreatic enzyme activity. Pancreatic enzyme products that are enteric-coated rely on exposure to pH>5.5 or 6 for 10-45 minutes to dissolve and release active enzymes. The SmartPill™ GI Monitoring System (Smart-Pill Corporation; Buffalo, New York) is an ambulatory wireless motility capsule (WMC) that measures pH, pressure and temperature of the GI tract and records the data on a receiver worn by the subject. We aimed to describe the gastrointestinal pH profile (GpHP) in patents with CF; to compare the GpHP of patients with CF to those without CF; and to determine if the GpHP of patients with CF is reproducible when measured at two points in time approximately one to two months apart. We hypothesize that time to neutralize gastric acid and sustain physiologic small bowel pH in CF subjects will be longer compared controls. Initial results are presented here.

Methods: A single-site, pilot study was conducted employing a convenience sample of 10 adult subjects with CF and 10 healthy subjects matched by age, gender and body mass index. Eligibility criteria specified that subjects could not be taking acid-suppressing medicines. Subjects on cycled inhaled antibiotics were mid-cycle during study periods. Eligible subjects ate a standard meal bar, swallowed the SmartPill™,and then were sent home. When the receiver indicated that the WMC had passed, subjects mailed the receiver back to the clinic where data was downloaded. Subjects were instructed to flush the WMC upon evacuation with stools. The first two hours of small bowel pH (5 second intervals) were analyzed and plotted for each subject. Identification of gastric emptying was indicated by an abrupt rise in pH of 3 or more units. The elapsed time from gastric emptying until a sustained (15 minutes) target pH above 5.5 and 6 was quantified using predicted values from the fitted local regressions. All statistical analyses were carried out using SAS version 9.1.3 statistical software (Cary, NC).

Results: Data from 7 CF subjects (2 studied twice), 2 controls (1 studied twice) and 10 historic controls are reported here. The CF subjects had a longer time to achieve pH 5.5 and 6. There was also a greater variability among the median, lowest and highest quartiles in CF subjects (See Table) .

Conclusion: In this preliminary report, subjects with CF were found to have an acidic duodenum for a longer period of time compared to healthy subjects, which is likely to impact the dissolution of enteric-coated PERT and contribute to malabsorption. Full data set will be presented at the conference.

Supported by a grant from CF Foundation Therapeutics, Inc. Background: Low serum levels of linoleic acid (LA) in cystic fibrosis (CF) patients have been attributed to malabsorption of fatty acids and an essential fatty acid deficiency. We have shown that LA levels are decreased in both human airway epithelial cells in culture and in UNC exon 10 transgenic CF mice due to increased conversion of LA to arachidonic acid (AA). This increased AA led to elevated downstream proinflammatory mediators and worsening lung inflammation. A high LA diet (6-12% of total calories) is the current recommendation for CF patients. We tested the hypothesis that high levels of dietary LA in patients with CFTR dysfunction leads to increased AA levels.

Methods: Healthy controls (HC), heterozygous (HT) and homozygous (CF) for CFTR gene mutations as confirmed by genotyping, were enrolled. All subjects met with a nutritionist. At baseline, at 3 weeks following a 100 gram fat/day with 1% LA and then for 3 weeks following an isocaloric diet with LA increase to 12%, samples were obtained including blood for serum and mononuclear cells, and nasal epithelial cells. Fatty acids were quantitated following extraction and analysis by GC/MS. Results: We report interim results of our ongoing study. Four HC (mean age 41), 5 HT (mean age 52) and 4 CF (mean age 17) subjects were enrolled. Shown in the table is the AA ratio and AA/LA ratio which compares these fatty acids measured on the high LA diet versus the low LA diet for each subject with increasing values reflecting increased AA production. The AA ratios in nasal epithelial and mononuclear cells were greatest in CF subjects compared to HT and HC subjects. In contrast, serum did not show any changes in the AA ratios. A greater fold increase was seen in the AA/LA ratios in CF subjects, consistent with LA levels not significantly increasing in CF subjects due to its enhanced conversion to AA. Intermediate AA/LA values in the obligate heterozygotes suggest a gene dosing effect.

Conclusions: These preliminary results demonstrate that a high LA diet can lead to increased AA levels in CFTR expressing cells in CF patients. The lack of changes in serum indicates that increased level of AA seen in mononuclear cells are not simply a reflection of the fatty acid environment in the serum, but rather a specific defect in fatty acid metabolism in the CFTR cells. Further enrollment and measurement of inflammatory markers are ongoing. Supported by the CF Foundation.

Dutta, A. 1 ; Woo, K. 1 ; Kresge, C. 1 ; Esser, V. 2 ; Feranchak, A. 1 1. Pediatrics, Univ Texas Southwestern, Dallas, TX, USA; 2. Internal Medicine, University of Texas Southwestern Medical Center, Dallas, TX, USA In the liver, biliary epithelial cells (BECs) are critical contributors to bile formation through regulated secretion of electrolytes and water. CFTR is located on the apical membrane of BECs and represents a pathway for Clefflux. However, the findings that i) only 20% of patients with cystic fibrosis (CF) (and non-functioning/absent CFTR on BECs) develop liver disease, and ii) Ca 2+ -activated secretion is 2 to 3-fold greater than that mediated by increases in cAMP via CFTR, suggest that alternate Clsecretory pathways exist in BECs and are important mediators of bile formation. However, the molecular identity of the Ca 2+ -activated Clchannel in biliary epithelium has not been defined. TMEM16A is a newly identified protein capable of transporting Clin other epithelium. The Aim of the present studies therefore was to determine if TMEM16A represents the channel responsible for Ca 2+ -activated Clefflux in BECs.

Methods: Studies were performed in Mz-cha-1 human BECs, isolated mouse BECs, or polarized rat biliary monolayers (NRC). Membrane Clcurrents were measured by whole-cell patch clamp techniques and reported as maximum current density at -80 mV. Transepithelial secretion was meas-ured by Ussing chamber and reported as change in short-circuit current response from basal (Isc).

Results: TMEM16A mRNA was detected by RT-PCR in all BEC models. Exposure to UTP rapidly increased [Ca 2+ ]i and activated whole-cell currents in both human and mouse BECs (-7.3±1.1 pA/pF and -33.9±0.3 pA/pF, n=2-4, respectively). Currents had a reversal potential at 0 mV (ECl=0) and mild outward rectification. UTP-stimulated currents were significantly inhibited by TMEM16A siRNA in both human (2.0±0.7 pA/pF) and mouse (15.5±3.2 pA/pF) BECs, but unaffected by mock transfection. In polarized NRC, UTP increased Isc (∆+24 ± 3 µA from basal) which was inhibited by the Ca 2+ -activated Clchannel blocker, niflumic acid (∆-16 ± 5 µA from basal), and increased by the TMEM16A potentiator IL-4 (∆+35 ± 6 µA). Together these results demonstrate that TMEM16A is functionally present in BECs and contributes to Ca 2+ -activated Clsecretion in response to extracellular nucleotides. These studies represent the first molecular identification of an alternate, non-CFTR, Clchannel in the liver and therefore may be a powerful target to modulate bile formation in the treatment of liver disorders such as cystic fibrosis. *Studies supported by the CFF (FER-ANC08G0). 

Background: Cystic fibrosis (CF) patients are considered common asymptomatic carriers of Clostridium difficile. At the Intermountain Pediatric CF center, Clostridium difficile-associated disease (CDAD) is more common than once thought. Recent literature has suggested acid suppression therapy may be a risk factor for CDAD in both hospital and community settings. To our knowledge, this association has never been studied in the CF population. The purposes of this study are to determine whether acid suppression therapy increases the incidence of CDAD, and to identify other risk factors for CDAD.

Methods: A retrospective chart review was performed using Cystic Fibrosis Foundation Patient Registry, Intermountain Primary Children's Medical Center microbiology lab data, and the electronic medical record. Patients who tested for C. difficile toxins between January 1, 2005 and December 31, 2008 were included. Demographic data such as age, gender, weight, genotype as well as the C. difficile toxin status were recorded. The type, use, and dosage of the acid suppression therapy (i.e. H2 blocker, proton pump inhibitor), and the receipt of antibiotic therapy (inpatient or outpatient) within the past three months were also collected. The primary outcome was the proportion of CF patients receiving acid suppression therapy who developed CDAD compared to those who did not receive acid suppression therapy and developed CDAD. Fisher's Exact Test was used to compare the primary outcome. Secondary endpoints were to determine other risk factors for acquiring CDAD. Mann-Whitney U test was used for secondary outcomes. This project was IRB-approved.

Results: A total of 189 cases were tested for C. difficile toxin. Sixtyeight duplicated cases were excluded. One hundred-seven (88%) of patients received a PPI, H2RA, or both and 14 (12%) did not receive any acid suppression therapy. CDAD developed in 62 of the 107 cases that received acid suppression and 7 of the 14 cases that did not receive acid suppression (p=0.58). Secondary outcomes will be presented.

Conclusion: The rate of CDAD between patients receiving acid suppression and patients who were not on acid suppression was not statistically significant. This might be due to insufficient power to detect such difference since the study did not include enough patients in the control group. However, all of the patients tested were symptomatic of CDAD in some manner (increased diarrhea, malabsorption, poor response to enzyme adjustment), and this appears to be a more significant problem than asymptomatic carriage. Future directions would be to look prospectively at CDAD in patients with acid suppression, response to CDAD treatments, the ability to clear the organism, and morbidity.

Robberecht, E.; Defreyne, L.; Voet, D.; De Bruyne, P.; Boelaert, H. Cystic Fibrosis Centre, UZ Ghent, Ghent, Belgium Background: Although bleeding from oesophageal varices is regarded as the main complication of portal hypertension (PH) due to cystic fibrosis (CF) related liver disease, increasing splenic enlargement often causes earlier concern. When the spleen becomes voluminous it provokes abdominal discomfort, appetite loss by gastric compression and hypersplenism with worrying thrombocytopenia and anemia. In general it begins within a few years after the occurrence of liver fibrosis (LF) and can rapidly progress as no treatment is available to stop it. Eventually splenectomy with surgical shunting can be performed but this procedure has serious risks, especially in those under 10 years. For all these reasons we investigated the potential of preventive TIPSS, before variceal bleeding, in 5 cf patients under 10y. after the early detection of PH.

Discussion: Liver abnormalities were detected at an early age (M4.9y; range: 4 -7.5y): LF by means of annual ultrasound (US) and in recent years also by fibroscan, and beginning splenomegaly, as a sign of PH, by systematic clinical examination. When the abnormalities increased over the following years TIPSS was installed after 1.3 -4.5y (M 2.5y) in an attempt to arrest further progression of PH. Follow up during the first six months after the procedure showed no improvement of established abnormalities like spleen enlargement or hypersplenic activity. On the other hand however deterioration was neither noted in any patient. This suggests that TIPSS can at least in the first six months after its installation, stabilise the effects of PH on the spleen. It can thus be of prophylactic value if applied soon after the detection of progressive LF, before the development of negative consequences of PH. It is however worrisome that it caused a persistent deterioration of biochemical markers for hepatic synthetic function in all patients. No other side effects were noted.

Conclusion: If long-term follow up confirms the preliminary findings, TIPSS can temporarily arrest the development of consequences from PH if timed early, before splenic enlargement and disturbances of hepatic function are obvious. Medical College of Wisconsin, Milwaukee, WI, USA Background: Self-assessment of sexual maturity in healthy adolescents has been shown to agree well with physician ratings. However, the high prevalence of delayed and attenuated pubertal growth (37% in our recent analysis), as well as altered body image in adolescents with CF, may affect the validity of self-assessment.

Objective: To examine the agreement between self-and physicianassessed sexual maturation in adolescents with CF.

Methods: We utilized data from 102 CF children enrolled in the Wisconsin Randomized Clinical Trial of CF Newborn Screening. Sexual matu-Results: Eight children (6 male, 2 female 9-17y, FEV1: 86%, 5 PI, 2 cirrhosis) responded to inclusion criteria and all wished to participate. At V1: pain intensity varied from mild (3) to severe (3), associated with chest (1) or head (2) pain, located mainly in the epigastric region (5) or left iliac fossa (5) , occurring almost daily (3) with a duration over 2 hours in 2 cases; high level of anxiety was present in 4 patients. Diary data correlated with the above pain characteristics in all but one child who presented no bouts of pain. One child asked to be withdrawn from the study. Pain management consisted in offering a change in schooling (1), dietetic (3) and PERT modifications (2) and in 3 patients additional training in hypnosis and autohypnosis. At V2, 6 months later: 2 had complete pain relief after stopping fizzy beverage or changing school. One child moved. In the remaining 4, frequency, intensity and duration of pain decreased dramatically, as well as emotional disorders and QoL improved.

Conclusion: This study is the first one asssessing prospectively RAP in CF children; we found a very low prevalence (6%) of RAP conversely to retrospective studies which report a high rate of abdominal pain "complaints." We suggest that ruling out trivial and transient disturbances, only a minority of CF children presents RAP criteria with a real negative impact on daily life. In those children, behavioral interventions working on anxiety can have a positive impact on pain relief, emotional disturbances and quality of life. We emphasize the need for further studies including larger cohorts. This work was supported by a grant from the Fondation de France. Objectives: To investigate efficacy and safety of long-term daily oral administration of a water-soluble new chemical entity of vitamin E (vE)tocofersolan (d-α-tocopherol-polyethyleneglycole-1000) in pediatric cystic fibrosis (CF) patients.

An open-label study was conducted in 32 pancreatic insufficient CF children followed in 5 French CF centers. Patients were treated with pancreatic enzymes and vE supplementation at a mean (SD) dose of 10.4 (4.3) mg/kg/d. At inclusion, they were switched from previous vE to tocofersolan at a dose of 6.7 mg/kg/d (derived from the recommended dosage in CF: 10 mg/kg/d synthetic α tocopherol, d-α-tocopherol being 1.5 times more potent). Efficacy was monitored by tocopherolemia (HPLC) and ratio of vE/lipids at baseline and after 3 and 6 months of treatment with tocofersolan. Safety analysis combined description of adverse events (AE), vital signs and laboratory safety data including vitamin A plasma level (vA). Statistics included Student's paired t-tests of timed end-points vs. baseline with error set at α< 0.05.

Results: Among 32 children included (18 male, 14 female) of median (min-max) age: 65.7 months (10. 3-154.5) and weight: 19.2 kg (7.1-43.3) , 29 completed the study (3 drop-outs: 1 consent withdrawal, 1 hair loss, 1 behaviour disorder). Compliance was excellent. At 3 and 6 months, plasma concentrations of vE (Table) , ratios of vE/lipids did not change significantly from baseline (under liposoluble or water-miscible preparations of dl-alpha-tocopherol). Tocopherolemia and ratios over lipids remained within respective normal ranges (12-35 µmol/L and >0.6 mg/g) and vA was unchanged after 6 months. Five serious AE (SAE) were recorded, 4 in 3 subjects during the first 3 months (bronchitis (2), failure to thrive, hair loss) and 1 during the last month (behaviour disorders). The relationship between tocofersolan and moderate hair loss, which completely recovered after cessation of the drug, is unknown, yet all other SAE were not related to the study drug.

Conclusions: Efficacy of tocofersolan, consistently maintained over time with normal indices of tocopherolemia, was similar to that of other formulations used in CF children. This is in accordance with data from a pharmacokinetic comparative study in CF demonstrating similar oral absorption of water-miscible vE and tocofersolan (Jacquemin E et al. J Clin Pharm Ther 2009; in press) . Tocofersolan was well tolerated at the dose regimen investigated during a 6 month-exposure, and represents a liquid preparation of vE ration ratings, i.e., Tanner stages, were obtained by physician examination and self-assessment in children older than 9 years of age beginning in 2001. Tanner stages are defined by pubic hair growth (PH) and genital (G) development in boys and PH and breast (B) development in girls, on a scale of one (pre-pubertal), two (onset) to five (fully mature). Kappa statistics was used to assess the agreement.

Results: Of 102 subjects, 89 (87%) performed at least one self-assessment (403 observations total); 85 (83%) were examined by physician at least once (350 obs); and 80 (78%) had at least one assessment completed by both self and physician (336 obs). The agreement between self-and physician-assessment at each Tanner stage is shown in the Table. In boys, for PH stages, 63% of observations were in agreement, 35% differed by one category. For G stages, 53% of observations were in agreement, 40% differed by one category. In girls, for PH stages, 83% of observations were in agreement, 16% differed by one category. For B stages, 83% of observations were in agreement, 14% differed by one category. All Kappa statistics were >0.7, indicating moderate to good agreement between self-and physicianassessment.

The agreement between self-and physician-assessment of sexual maturation is good, and better in girls than boys. Further analysis examining the consistency of self-assessment and the relationship of pubertal onset to adolescent growth acceleration will be presented.

(Supported by NIH R01DK072126 & R01DK34108.) Chronic pain in children with cystic fibrosis (CF) is reported as being mainly abdominal with a frequency up to 30%. Unrecognized and untreated it reduces quality of life and ability to participate in CF-related daily care.

Aims: 1) Assess prospectively the incidence and the characteristics of recurrent abdominal pain (RAP) defined by "at least three bouts of pain, severe enough to affect activities, over a period of not less than three months, with attacks in the year preceding the examination" (Apley J. Arch Dis Child1958); 2) determine the cause of pain and provide pain relief, with an assessment 6 months later.

Materials and methods: One hundred-thirty patients (8-18y) from a CF center, which is part of a GI unit, were systematically questioned on RAP and were offered to join the study. Initial visit (V1) included: clinical and paraclinical assessment to diagnose the presence of an organic disease, pain location (Eland), pain intensity (Faces Pain Scale-Revised), emotional status (Mc Gill Qo) and anxiety levels (R-CMAS 1999) , quality of life (CF-QoL, Henry 2003) . A prospective 28 days pain survey diary was given to the patient. A synthesis meeting (referent physician, psychologist, pain team, and family) analyzed individual data and the pain management program was offered, including if necessary behavioral interventions (like hypnosis). Six months later (V2) a similar complete re-assessment was performed. and prolonged. Although there are several eponymous regimens advocated for its treatment, we had the impression that the management of DIOS was not uniform across the CF caregiver community. To look at this further, we sought the opinions of 23 CF centers within the UK (catering for a total of 4388 people with CF) regarding their approach to the diagnosis, treatment and prevention of DIOS in their CF population. Overall, 211 patients in these centers had suffered this CF complication in the last year, with 1 center alone reporting 40 cases.

As regards diagnosis, all centres included abdominal pain and definitive bowel obstruction, but some included an altered bowel habit and nausea and vomiting.

However, when patients presented acutely with bowel obstruction, only 16 centers (70%) had a written policy for its treatment. The most common initial therapy was oral gastrograffin (18 centers, 78%), followed by lactulose and movicol (both 2 centers, 9%). Gastrograffin use varied widely in terms of amount and frequency.

Although all centers would subsequently treat resistant cases with oral bowel flushing agents based on a physiological solution (e.g. "Klean Prep"), this was only administered by nasogastric tube routinely in 13, with 1 center using this routinely at first presentation. Twelve centers (52%) sought advice from surgical teams for resistant cases, but DIOS played a part in the subsequent death of only 3 patients.

In terms of ongoing management once the acute episode had resolved, all centers gave advice on the early recognition of symptoms, hydration, diet, and enzyme control, and 21 centers (91%) prescribed regular laxative therapy to diminish the possibility of recurrence.

The majority of the centers questioned had a multidisciplinary team approach to DIOS, including input from CF specialist medical staff, nurses and dieticians, but in 1 center (4%) patients were handed over to the care of a gastroenterology team.

Although a seasonal variation in DIOS is described, only 8 centers (35%) had noted an increase in cases during the summer months, possibly related to the many wet summers recently experienced in the UK.

From this work it is apparent that there is no uniform policy on the treatment and prevention of DIOS in adult CF patients in the UK, despite the presence of several protocols. Further work needs to be done to establish the most appropriate way of managing this unpleasant complication of CF which occurs with increasing frequency in a subgroup of adult people with CF as they become older. Background: A strong association exists between CFTR genotype and exocrine pancreatic phenotype, with mild mutations conferring pancreatic sufficiency in a dominant fashion. Pancreatitis occurs in approximately 15% of cystic fibrosis (CF) patients with pancreatic sufficiency (CFPS-PANC) and in a subset of patients who carry CFTR mutations but do not fulfill the diagnostic criteria for CF (CFTR-related pancreatitis; CFTR-PANC). In our experience, patients with pancreatic insufficiency do not develop pancreatitis. Our aim was to evaluate the association between severity of CFTR mutations and the development of pancreatitis by comparing patients with CFTR-PANC, pancreatic sufficient CF with pancreatitis (CFPS-PANC) and without pancreatitis (CFPS-NOPANC).

Methods: Genotype and clinical data of patients with CFTR-PANC, CFPS-PANC and CFPS-NOPANC were ascertained from Toronto, Verona and the Canadian Consortium for CF Genetic Studies. Only CFTR-PANC patients who carried 2 CFTR mutations were included for analysis. The predicted functional severity of CFTR mutations was determined by the Pancreatic Insufficiency Prevalence score among >2500 Canadian CF patients with defined pancreatic function status. Mutations were classified based on the proportion of patients with pancreatic insufficiency for each specific mutation: mild (≤0.25), intermediate (0.26-0.74) and severe (≥0.75). The risk of pancreatitis, age at onset of pancreatitis, gender, sweat chloride and FEV1 (% predicted) were compared between the 3 patient groups. Background: Our recent analysis demonstrated that cystic fibrosis (CF) infants exclusively breastfed for one month or longer exhibited lower weight and length z-scores during the first 2 years of life compared to those exclusively formula-fed. Inadequate pancreatic enzyme replacement therapy (PERT) may contribute to the differences in growth outcomes.

Objective: To characterize PERT dosing, and their associations with growth, in breastfed and formula-fed CF infants during the first 2 years of life.

Methods: The study population includes a subgroup of 70 CF infants enrolled in the Wisconsin Routine CF Newborn Screening Study who were born during 1994-2006, had no meconium ileus and no mutations known to be associated with pancreatic sufficiency. Extensive medical record review was conducted to obtain data on infant feeding and PERT at all clinic contacts from diagnosis (range 0-2.4 wk) to age 2 years. Infants were classified into 4 feeding groups: exclusively breastfed <1 month (ExBF1mo-, n=16), exclusively breastfed ≥1 mo (ExBF1mo+, n=18), exclusively fed formula containing 20 kcal/oz (ExFM20, n=22), and exclusively fed formula containing ≥22 kcal/oz (ExFM22+, n=14).

Results: Of the 54 infants (77% out of 70) who received PERT, average lipase dose (units/kg/feeding) increased significantly with age (514±47/year, p<0.001) in all feeding groups. However, average lipase dose was lowest in the ExFM20 group (1339±716), which was significantly lower than the ExFM22+ group (1968±818, p=0.010) and the ExBF1mo-group (2251±754, p=0.013), but was only borderline significant compared to the ExBF1mo+ group (1814±838, p=0.082). Weight gain during the first two years of life remained significantly lower in the ExBF1mo+ group than the ExFM20 group, p=0.045, and lipase dose was found to modify this association. However, the association was complex; a significant interaction (p=0.019) between feeding group and lipase dose was observed. In the ExFM20 and BF1mo-groups, lipase dose was not associated with weight gain (p=0.91 and 0.99, respectively). On the other hand, lipase dose was positively associated with weight in the ExBF1mo+ group (p=0.049) but negatively associated with weight gain in the ExFM22 group (p=0.026). More specifically, in the ExBF1mo+ group, infants receiving lower PERT (<1500 units/kg/feeding) had significantly better weight gain compared to infants receiving higher PERT (>1500 units/kg/feeding), p=0.034. In the ExFM22 group, infants receiving higher PERT (>1500 units/kg/feeding) had significantly better weight gain than those receiving lower PERT (p=0.056).

Conclusion: Complex associations among PERT, breast/formula feeding, and weight gain were discovered. Nevertheless, weight gain during the first two years of life remained significantly lower in CF infants exclusively breastfed longer than one month compared to those exclusively formulafed after adjusting for PERT dosing. Further analyses examining how PERT varies with the duration of breastfeeding will be presented. The distal intestinal obstruction syndrome (DIOS) is a complication of the cystic fibrosis (CF) condition which increases with age, and occurs in up to 20% of adult patients. In severe cases, its treatment can be problematic Results: Of 131 patients, the prevalence of vitamin D insufficiency at baseline was 67% (39% mild, 23% moderate and 5% severe). Of these 88 patients with vitamin D insufficiency, 11 (13%) did not receive vitamin D supplementation and 14 (16%) did not have follow-up serum levels. Of the remaining 63 patients who received supplemental vitamin D and had at least one follow-up serum 25(OH)D measurement, serum 25(OH)D at first follow-up (6.4 ± 4.4 months after supplementation) improved from 20.0 ± 5.8 ng/ml at baseline to 30.0 ± 12.8 ng/ml at follow-up, p < 0.001. In addition, serum 25(OH)D improved in most (84%) of these 63 patients; approximately half (51%) improved to >30 ng/ml and the other half (49%) increased, but did not reach 30 ng/ml. At the first follow-up, 36 patients (57% of 63) remained vitamin D insufficient (40% mild, 14% moderate and 3% severe). Adherence to vitamin D supplementation, subjectively reported by care providers as good in 43%, moderate in 43% and low in 14% of the 63 patients, was not associated with treatment response (p=0.69).

Conclusion: In the UW Pediatric CF center, two-thirds of CF children were vitamin D insufficient before implementing the comprehensive vitamin D supplementation protocol. Serum 25(OH)D was normalized in half of vitamin D insufficient patients after receiving vitamin D supplementation for an average of 6 months. Further analysis comparing the effectiveness of cholecalciferal (vitamin D3) versus ergocalciferol (vitamin D2), and the dose and duration of supplementation, on treatment response will be presented.

(Supported by NIH-R01DK72126 and 1MCHB 5T72MC00008-18-00.) Roddy, M.F. 1, 2 ; Clancy, G. 2 ; Leen, G. 2 ; Elnazir, B. 2 ; Greally, P. 2 1. Dietetics, AMNCH, Dublin, Ireland; 2. Cystic Fibrosis, AMNCH, Dublin, Ireland Marie Roddy1, Geraldine Clancy2, Geraldine Leen2, Basil Elnazir2 and Peter Greally2. 1 Department of Clinical Nutrition and Dietetics, 2 Cystic Fibrosis Department, Adelaide and Meath Hospital, Incorporating The National Children's Hospital (AMNCH), Dublin Introduction: Determination of pancreatic function in children who have cystic fibrosis (CF) is essential to optimize health, improve clinical outcomes and avoid unnecessary pancreatic enzyme replacement therapy (PERT). The 72 hour faecal fat analysis is routinely used to diagnose pancreatic function but this test is time consuming and intrusive. We hypothesised that the simple non-intrusive stool elastase test may be equivalent or even more accurate than the 72-hour faecal fat analysis.

Methods: A retrospective review of pancreatic function tests (72 hour faecal fat analysis and stool elastase) was conducted. Both tests had been carried out at the same time for each patient. The 72-hour faecal fat analysis was analysed at Claymon Laboratories while a 3-day food diary kept during the time of stool collection was analysed using Dietplan®. A result of <93% co-efficient of fat absorption (CFA) was classified as Pancreatic Insufficient (PI). Stool elastase was measured using the ELISA monoclonal method (Schebo, Tech, Wettenburg, Germany). A level of < 200µg/g was classified as PI using this method. The cost of the stool elastase test is 28 Euro* while the 72 faecal fat analysis costs 58 Euro per test*. Minitab Statistical Package was used to analyse the data. Regression analysis was carried out to determine the level of correlation between the 2 methods as well as Chi Square analysis.

Results: Fifteen children (9 females) with mean age 3.6± 0.77y (SE) were included in the study. Seven (47%) were classified as PI using the 72hour faecal fat analysis but this increased to ten (67%) when the stool elastase method was applied. These further three patients labelled PI by stool elastase were displaying signs of malabsorption and subsequently did well on PERT. The stool elastase did not significantly correlate with the 72 hour faecal fat analysis (CFA), r=0.46 (p=0.08). Chi Square also showed that Stool Elastase and the 72 Hour faecal fat analysis gave different results (χ2 = 6.56, p= 0.01). There was also no correlation between the stool elastase and amount of fat/ 100g of stool (r=-0.38, p=0.16) or daily fat excretion (r=-0.39, p=0.15). However stool elastase did correlate significantly with faecal weight (r=-0.56, p=0.031).

Conclusion: These results highlight the difficulty in the diagnosis of pancreatic function in children with CF. The 72 hour faecal fat analysis is fraught with potential errors as it depends on adequate collection of stool and recording of dietary intake by patients. In conclusion, we feel the stool elastase is an accurate, non-intrusive, simple and cost effective method of determining pancreatic function.

*cost on 22nd April 2009 and VAT included. Cystic fibrosis transmembrane conductance regulator (CFTR) is a chloride channel located on the apical side of epithelial cells lining the gastrointestinal tract. CFTR mutations cause cystic fibrosis (CF), a disease characterized by recurrent pulmonary infections, pancreatic insufficiency, intestinal malabsorption and steatorrhea. Remarkable benefits resulting from exogenous pancreatic supplementation have contributed to the continual improvement of the quality of life of CF patients. However, despite these significant steps forward, fat malabsorption remains a persistent feature in CF patients. It has been previously shown that these patients exhibited abnormalities in the intra-enterocyte lipid processing. Studies aimed at increasing our understanding of the intestinal handling of macro-and micronutrients in CF are multiple although none of them has specifically addressed the direct implication of CFTR in lipid absorption and metabolism by intestinal cells. We carried out studies aimed at investigating the role of CFTR in intestinal synthesis, lipid transport and secretion using Caco-2/15 cell line.Using shRNAi, CFTR knockdown in Caco-2/15 cells resulted in the reduction of gene (70%) and protein (50%) expression. CFTR invalidation led to a significant increase in lipid synthesis and secretion, which was mainly attributed to raised levels of phospholipids, triglycerides (TG) and cholesteryl ester. In line with the augmented level of TG, an induction was noted in the activity of monoacylglycerol and diacylglycerol acyltransferase. Analysis of lipoprotein metabolism showed an enhanced secretion of TG-rich lipoproteins, as well as low-and high-density lipoproteins, including most of their apolipoproteins. Accordingly, the activity of microsomal transfer protein, a key protein in lipoprotein assembly, was elevated. Moreover, the total amount of fatty acids (FA), especially of saturated FA, was significantly higher in cells with CFTR knockdown, confirming the magnitude of lipid absorption. On the other hand, the expression of liver-fatty acid binding protein, a cytosolic FA binding protein, was strongly induced following CFTR knockdown while phosphorylation of acetylCoA carboxylase was significantly reduced, thereby allowing more lipogenesis to occur. Finally, investigation of cholesterol (CHOL) metabolism revealed that CFTR knockdown had no impact on [ 14 C]-CHOL uptake by the apical membrane of Caco-2/15 cells and on the level of proteins involved in CHOL influx [Scavenger Receptor Class B Type I and Niemann-Pick C1 like 1] and efflux [ATP-binding cassette G8)]. Overall, these data demonstrate that the reduction in CFTR expression alters intestinal FA metabolism and lipid transport in intestinal absorptive cells. Our findings may suggest that CFTR defects are not directly responsible for the abnormal intracellular phase of lipid transport in CF and that other confounding factors (essential FA deficiency, intraduodenal acidification) might explain the abnormal intra-enterocyte events contributing to fat malabsorption in CF patients. Wisconsin, Milwaukee, WI, USA; 5. Akron Children's Hospital, Akron, OH, USA; 6. American Family Children's Hospital, Madison, WI, USA; 7. Texas Children's, Houston, TX, USA Background: Up to 90% of cystic fibrosis (CF) patients have exocrine pancreatic insufficiency, resulting in poor absorption of nutrients including fat-soluble vitamins A, D, E, and K. There is also evidence that CF patients who are pancreatic sufficient may have increased vitamin requirements. Research is constantly emerging on the role of vitamin D in pulmonary health and the roles of vitamins A, D, and K in bone health. A line of CFspecific multivitamins (SourceCF) was designed for CF patients and provides a range of formulations for all ages. In recent years, new data have been published that provide evidence that vitamin formulations could be further optimized.

Methods: The development of SourceCF multivitamins in 2003 was guided by the CFF Consensus Report on Nutrition for Pediatric Patients with CF, the CFF Adult Care Consensus Conference Report, published research, interviews with international vitamin experts, and input from the SourceCF Nutrition Advisory Council (SNAC; consisting of 16 RDs from 16 US CF centers). In 2008, SNAC performed an extensive review of CF literature published from 2003 to 2008 that focused on vitamin research. SNAC also reviewed vitamin evidence for the general public. Based on the findings, SNAC developed recommendations for the reformulation of SourceCF multivitamins.

Results: Thirteen studies with a total of 1456 CF subjects were identified, providing evidence that CF patients have greater requirements for vitamin D (5 studies; n=1042) and vitamin K (6 studies; n=263) compared with recommendations prior to 2003. One study found 70% of children with CF in pancreatic enzyme supplementation, aggressive nutritional interventions, and improvements in the treatment of infection and inflammation have led to improved but not fully corrected body mass index profiles in CF patients. The CF mouse also exhibits growth restriction, has intestinal obstruction and typical CF epithelial ion transport characteristics without the confounders of exocrine pancreatic dysfunction or spontaneous pulmonary infections. These observations suggest that there may be additional factors contributing to growth retardation. Because CFTR dysfunction in the intestinal epithelium is implicated in causing intestinal obstruction it was theorized that it could also cause malabsorption and subsequent growth retardation.

Study of growth in CF mouse models is difficult because sublethal obstructive events inhibit growth. Diets with laxatives limit obstruction but prevent adequate study of absorption. A murine allele of the CF missense mutation R117H was studied to address these potential confounders. In humans this mutation is associated with milder disease and pancreatic exocrine sufficiency. R117H mice experience fewer lethal intestinal obstructions (<10%) than cftr null mice (70%) but still show growth restriction similar to other CF mouse models.

Mice were fed a diet containing the non-absorbable biomarker, sucrose polybehenate, and gas chromatography was used to measure ratios of this marker to other fecal fatty acids. There was no difference in fecal fat between S489X CF mice (n=7, 93.8% absorption), S489X wild type control mice (n=6, 97.7% absorption) and "gut-corrected" (FABP-CFTR) mice (n=3, 96.5% absorption). Additionally, fecal fatty acid profiles were determined by mass spectrometry. There were no detectable differences in the fecal fat content of wild type (n=4) and R117H CF mice (n=4). In comparison to the chow, nearly all of the essential fatty acids were absorbed by both CF and wild type mice.

In CF patients, whose malabsorption is corrected with pancreatic enzyme supplementation, growth deficits are improved by implementation of a high fat, high caloric diet. We attempted to mimic this practice by creating a hyperphagic CF mouse using a leptin receptor mutation. The ∆F508, leptin receptor double-knockout mice did become fat but were still short. Lethal obstruction prevented further study and therefore a leptin receptor, R117H double-knockout mouse is being created to further explore if increasing caloric intake can improve growth in the CF mouse.

Pancreatic insufficiency, lung infections, intestinal obstruction, and malabsorption do not cause growth retardation in the R117H CF mouse. The primary cause of growth retardation in the CF mouse and whether or not this can be overcome with increased caloric intake remains unknown. Determination of the cause of growth retardation, and the best way to improve it, has great importance to patients whose mortality remains linked to their body mass.

O'Brien, C.E. 1 ; Anderson, P.J. 2 ; Stowe, C.D. 1 Background: Constipation is common among adults with cystic fibrosis (CF). Lubiprostone is a medication indicated for the treatment of idiopathic chronic constipation in adults and irritable bowel syndrome with constipation in women. It works locally by activating type 2 chloride channels in the intestinal tract. Its activity is not dependent on a functional CFTR, and therefore may be an effective agent for the treatment of constipation in patients with CF.

Objective: Determine the efficacy of lubiprostone for the treatment of constipation in adults with CF.

Methods: This is a 4-week open label pilot study with a 2-week run-in period during which subjects use their current medication for constipation. During the 4-week treatment period subjects use lubiprostone 24 mcg twice daily. Efficacy measurements are taken before treatment ("baseline") and at 2 and 4 weeks of treatment. The primary efficacy endpoint is the number of spontaneous bowel movements (BM) per week as measured by a BM diary. Secondary efficacy endpoints include mean Bristol Stool Scale (BSS) score and Patient Assessment of Constipation Symptoms (PAC-SYM©1999 Johnson & Johnson) survey score. Measurements during the run-in period are compared to those during the treatment period by One Way ANOVA for repeated measures. Safety endpoints include pulmonary function tests (PFTs), CF clinical score, serum chemistry, and body mass index (BMI). PFTs and serum chemistry are measured before and after the 4-week treatment period and compared with paired t-tests. For all analyses, a P value of <0.05 is considered statistically significant.

Results: Five subjects have completed the pilot study to date (40% male). Baseline characteristics (mean ± SD) included: age 24.4 ± 3.7 years, FEV1 % predicted 84.6 ± 11, and BMI 23.8 ± 2.9. There was no statistically significant difference in BM frequency or BSS scores across the study period. There was a statistically significant improvement in PAC-SYM scores across the study period (Table) . There was not a statistically significant difference in safety measures before and after treatment with lubiprostone.

Conclusions: These preliminary data suggest that lubiprostone may be an effective option for the treatment of constipation in adults with CF. This study is ongoing.

Acknowledgments: Supported by a grant from Takeda Pharmaceuticals. In healthy adolescent females, the pubertal growth spurt begins with a large increase in linear growth velocity over the first year then decreased velocity as increased estradiol causes epiphyseal closing. Among non-CF females, peak height velocity occurs on average 0.5 years prior to menarche. We aimed to describe peak height velocity timing and magnitude in relation to menarche in an adolescent CF population to test the hypothesis that increased growth velocity is a reliable proxy of onset and progression of puberty.

Methods: A cohort of all 18-21 females with CF year olds was retrospectively identified from a single, large US CF Foundation accredited CF Center. Longitudinal height data from ages 10-18 were used to calculate peak height velocity based on the best possible approximation of a 12 month time period. Height velocity was calculated by taking the difference of height measurements with maximum value assigned as the peak height velocity (cm to 1 decimal point/year) for the individual. Menstrual data (months of age at menarche) was abstracted by retrospective CF chart review. The Pearson's correlation coefficient was computed to evaluate the linear relationship between the age of menarche and age at peak height velocity.

Results: The cohort included 26 females with CF with mean BMI % predicted 43 ± 28 and FEV1% predicted 82 ± 16. Median peak height velocity was 7 cm/year (range 1-10.1) and occurred at 12.0 years (range 11.0-15.9), compared with non-CF population norms of 8.3 cm/year at median chronologic age of 11.5 years (1). Two years of sustained height velocity was observed in 42% (11 of 26). Eighteen subjects had menarche data available with median age of menarche 13.0 years (range 9-16 years) vs. mean age of menarche in a white non-CF population at 12.5 years (1). There was no correlation between age of menarche and the age of peak height velocity among individuals in this cohort (Pearson correlation coefficient -0.0119,p-value 0.963).

Conclusions: Peak height velocity occurred later and with less robust velocity in females with CF when compared with population norms.

Bacon, E.A.; Acton, J.; VanDyke, R.; Fencel, M.; Boesch, R.P.; Seid, M.; McPhail, G. Pulmonary, Cincinnati Children's Hospital, Cincinnati, OH, USA Background: Patients with cystic fibrosis (CF) who are at nutritional risk may receive gastrostomy tubes (G-tube) as a definitive intervention to deliver enteric feedings and improve overall caloric intake.

Goals: The goal of this project was to determine if G-tube placement improved nutritional and lung function outcomes in our CF center. Our hypothesis was G-tube placement would be associated with improvement in nutritional outcomes but would not be associated with improvement in lung function.

Methods: We queried our local CF database for patients in our pediatric center who had G-tubes placed and seen in clinic in the last 15 years. Nutritional analysis included 45 patients and lung function analysis included 36 patients. Patients were 1 month to 16 years old. Data were analyzed to look for differences in outcomes modeling pre-insertion and post-insertion data. The data were analyzed using a repeated measures random-intercepts' model. Differences in least squares means were calculated. Modeling was performed for 7 different dependent variables and their slopes: weight-forage z-score, height-for-age z-score, BMI-for-age z-score, FEV1% predicted, and logit transformations of percentiles for weight-for-age, height-for-age, and BMI-for-age. Initial covariates included age, sex, genetics, race, diabetes status, and chronic infection status. Using backwards elimination, nonsignificant variables were removed from the final models.

Results: G-tube insertion was associated with a better weight-for-age zscore (mean increase of 0.24; p<0.0001) and better BMI percentile (mean increase of 0.29; p=0.014) when compared to pre-insertion status. G-tube insertion was associated with a worse rate of decline in FEV1% predicted (mean slope change -0.7% per year; p=0.002) and lower absolute lung function (mean FEV1% predicted 5.9% lower, p<0.0001) when compared to pre-insertion status. Older age at G-tube insertion was associated with better baseline nutritional status at time of insertion. However, older age at insertion was associated with a worse post-insertion slope for all nutritional variables and trended towards an association with a worse rate of decline in lung function after insertion (p=0.08). In patients with G-tubes, infection status was an important predictor of lung function and nutritional status. Patients with chronic infections with MRSA, B. cepacia, or mucoid P. aeruginosa either trended towards or had lower FEV1% predicted. Chronic infection with MRSA or MSSA was associated with a lower weight-for-age percentile and a lower BMI-for-age percentile.

Conclusion: This analysis showed G-tube placement is associated with improved nutritional outcomes. It also revealed that in this center, G-tube Although many subjects had less than expected yearly growth at the time of peak height velocity, some had sustained increases in growth velocity exceeding expected norms. Menarche was delayed when compared to non-CF cohorts. In our study, peak height velocity did not reliably correlate with menarche. Further research is needed to understand how delayed and possibly prolonged pubertal growth in CF affects pubertal development and whether this contributes to sustained metabolic demands and the gender gap. Vitamin D insufficiency is common in patients with cystic fibrosis (CF). Recent studies have demonstrated that vitamin D may be important for immune system by increasing production of cathelicidin, a potent antimicrobial peptide with activity against bacteria. Thus, rapid correction of vitamin D status may be important in CF patients admitted for acute respiratory exacerbation. The purpose of this study is to examine the safety and efficacy of a single large dose of vitamin D 3 in CF patients during a hospitalization for an acute respiratory exacerbation. Inclusion criteria include CF patient admitted for respiratory exacerbation and age >18. Exclusion criteria include current therapy with high dose vitamin D (>2000 IU QD) or an admission for serious terminal illness. The study was approved by the Emory IRB. Subjects were randomized in blocks of 6 to either 250,000 IU of vitamin D 3 or placebo given orally as a single dose within 48 hours of admission. Blood levels for 25-hydroxyvitamin D (25(OH)D) were drawn at randomization and at the time of hospital discharge. We report the results of the first 12 out of 30 subjects to be enrolled in this study. The subjects were a mean age of 28.5 ± 9 yrs. Vitamin insufficiency or low normal vitamin D status <40 ng/mL) was present in 92% of subjects. Subjects randomized to vitamin D 3 250,000 IU had a baseline 25(OH)D of 30 ± 14 ng/mL which increased to 51 ± 10 ng/mL (p=0.02) after a mean hospital stay of 8 ± 3 days. Subjects randomized to placebo had a baseline 25(OH)D of 30 ± 7 which was unchanged at 29 ± 10 (p=0.8) after a mean hospital stay of 8 ± 3 days. It is important to note that no subjects developed hypercalcemia.

In conclusion, a single large bolus dose of vitamin D 3 was able to rapidly restore vitamin D status in CF patients admitted with acute respiratory exacerbation. This study provides preliminary evidence for a vitamin D repletion regimen which results in optimal vitamin D status in hospitalized CF patients. Further intervention studies should be conducted to determine whether rapid correction of vitamin D translates into improved clinical outcomes such as improved respiratory and immune function. placement may be associated with a faster rate of decline of lung function. A Registry level analysis is needed to further evaluate nutritional and lung function outcomes after G-tube placement. Background: Gastrointestinal manifestations of cystic fibrosis (CF) are common and can be difficult to treat. Lactose malabsorption is often cited as a cause of gastrointestinal symptoms in CF. Lactose Intolerance occurs in approximately 15-20% of people of European descent. Lactose Intolerance in CF has been studied in the paediatric population but never in adult patients with CF.

Objectives: The purpose of this pilot study was to determine the prevalence of Lactose Intolerance in adult patients with CF.

Methods: 20 adult patients with CF completed a lactose-hydrogen breath test after an overnight fast. A positive breath test was defined as a rise of 20 ppm from baseline values with fasting breath hydrogen levels <20 ppm considered normal. Each participant completed an interviewer-assisted questionnaire to identify any history of distal intestinal obstruction syndrome (DIOS), frequency of gastrointestinal symptoms and use of laxatives.

Results: All patients were pancreatic insufficient (PI). One patient (5%) met the diagnostic criteria for Lactose Intolerance. Thirteen subjects (65%) had a fasting breath hydrogen ≥20 ppm indicating small intestinal bacterial overgrowth (SIBO). There was no association between fasting breath hydrogen level and reported gastrointestinal symptoms.

Conclusions: Lactose Intolerance is uncommon in patients with CF. Abnormal hydrogen breath tests occur frequently in CF with small intestinal bacterial overgrowth (SIBO) suspected in 65% of subjects based on raised fasting breath hydrogen levels. Background: Within the cystic fibrosis (CF) population approximately 10 -15% will be pancreatic sufficient (PS). The Cystic Fibrosis Trust advocates annual assessments for all patients (1) . Annual assessments should include an oral glucose tolerance test (OGTT) and measurement of fat soluble vitamin levels (1) .

We reviewed the OGTT and vitamin results of PS patients to assess whether this lead to a change in their care.

Method: We reviewed the most recent OGTT results and fat soluble vitamin levels from annual assessments of the PS patients within our clinic population. Results were obtained from the hospital computer results system, taking note of time of year vitamin levels were measured.

Results: Twenty-two patients with PS were identified. Of these, 20 had had an annual assessment within the last 2 years. Nineteen had a normal OGTT result (1 patient was known to have diabetes mellitus, thought to be Type 1 Diabetes Mellitus rather than CF Related Diabetes).

All patients had normal vitamin A and vitamin E levels (Vitamin A mean 1.72µmol/L, range 1.11 -2.45µmol/L; vitamin E mean 26.5µmol/L, range 21 -36.8µmol/L), none of whom were receiving vitamin supplementation.

Eighteen patients had a vitamin D result within the normal range for our hospital in accordance with the time of year vitamin D was measured (10 -30ng/ml in winter, 15 -60ng/ml in summer). Two patients had vitamin D insufficiency. Four patients were taking CalcichewD3Forte, 2 as bone protection as patients were taking long-term prednisolone, 2 due to patients having previously low vitamin D levels.

Ten patients (including 2 with vitamin D insufficiency) had a vitamin D level below 30ng/ml, as recommended by the CF trust to help prevent low bone mineral density (2) .

Discussion: These results suggest it may not be necessary for patients who are PS to have an annual assessment in its current format including fasting vitamin levels and OGTT. It may still be necessary to measure vitamin D and supplement as required.

References : 

fat absorption (CFA) following oral administration of EUR-1008 vs placebo while on a controlled diet. Patients were prohibited from taking concomitant treatments such as proton pump inhibitors (PPIs) or other medications that could affect gastric motility or pH. Results: Twenty of the 34 enrolled patients had been taking agents affecting gastric pH prior to enrollment, which they discontinued upon entering the trial. A post hoc analysis revealed that patients taking gastric acid modifiers prior to enrollment reported a greater improvement in symptoms of malabsorption on EUR-1008 than those who did not require PPI or H2 antagonists with their previous PEP treatment. Additionally, all patients, whether they had a prior history of gastrointestinal (GI) medication use or no prior GI medication use, required a similar dose of EUR-1008 for symptom management (mean 4629 ± 1995 U/kg/day for those without prior PPI or H2 antagonist use vs mean 4568 ± 1284 U/kg/day in those with prior PPI or H2 antagonist use).

Conclusions: Since the effects of EUR-1008 were obtained in the absence of any concomitant treatment affecting GI motility and pH such as PPIs (which most patients were taking before entering the trial), it appears that the efficacy of EUR-1008 is independent of their concurrent use. By eliminating these agents and decreasing pill burden, EUR-1008 may allow for improved enzyme regimen compliance, resulting in better overall nutritional status for CF patients with EPI. These findings need to be confirmed in clinical studies. Cystic fibrosis (CF) has been associated with abnormalities in the lipid metabolism specifically with reduced levels of fatty acids from the omega-3 family such as docosahexaenoic acid (DHA) and elevated concentrations of omega-6 fatty acids such as arachidonic acid (AA). We have found these abnormalities in our CF mouse model (C57BL/6J-Cftr m1Unc ) and in patients with CF. Recently, we have described in CF mice and patients defective levels of some species of ceramides. We also found the lipid abnormalities to be correctable by treatment with fenretinide, , a semi-synthetic retinoid. To determine the source of this lipid imbalance and to elucidate the mechanism of fenretinide's effects on fatty acids, we proceeded to determine whether the fatty acid imbalance depended on CFTR and whether various cell types would respond to the drug. CFBE cells demonstrated similar fatty acid abnormalities as seen in CF mice and patients. These aberrant fatty acid profiles were corrected by the transfection of wild-type CFTR, indicating that the fatty acid abnormalities are linked to the primary gene defect. All cells, transfected and untransfected, were treated with fenretinide. Both groups displayed an increase in DHA and ceramide, and decreased AA following drug treatment however the greatest effects were found in untransfected cells.

We isolated leukocytes from 10 CF patients which displayed low levels of DHA, ceramide and high levels of AA. Following treatment with fenretinide, DHA and ceramide increased (p = 0.0028 and p < 0.0001, respectively) and AA decreased significantly (p < 0.0001). These results indicate for the first time that fenretinide can act on a systemic level in CF patients. Interestingly, bone marrow macrophages from our CF mouse model also displayed similar fatty acid abnormalities (low DHA, high AA levels and low levels of ceramides) which were improved following treatment with feneretinide. We quantified the levels of ERK1/2 phosphorylation pre-and post-stimulation with lipopolysaccharide (LPS). As expected LPS stimulation increased the levels of ERK1/2 phosphorylation. Interestingly, when cells were treated with fenretinide, ERK1/2 phosphorylation was reduced to basal levels even in the cells treated with LPS. These results demonstrate an innate augmentation in ERK1/2 phosphorylation most likely linked to the aberrant fatty acid metabolism in CF cells. The majority of cystic fibrosis (CF) patients have exocrine pancreatic insufficiency (EPI), requiring supplementation with pancreatic enzyme products (PEPs). EUR-1008 is a novel, enteric-coated, zero-overfill PEP. Very young children and some adults have difficulty swallowing whole capsules. The ability to open capsules and disperse the contents onto foods is an important option for these patients; however, there are no published data on the effect of dispersing on the efficacy of PEPs. In order to examine whether dispersing affects the efficacy of EUR-1008, an exploratory analysis of patients administered EUR-1008 by dispersion of capsule contents onto foods vs those receiving it by capsule was performed.

Methods: This exploratory analysis of an open-label multicenter trial included children aged 1-6 years with CF and EPI, with good nutritional status and fecal elastase <100 µg/g. During the trial, patients consumed a standard recommended CF diet without concomitant agents that could affect gastric pH or motility. Patients were switched from their previous PEP to EUR-1008, with a 4-day screening period, a 7-day dose-stabilization period, and a 7-day treatment period. The primary efficacy end point was the percentage of responders, defined as patients with modest steatorrhea (<30% fecal fat content) and no signs or symptoms of malabsorption. Clinical symptoms as well as patient and caregiver assessment of signs and symptoms of malabsorption were recorded.

Results: Nineteen patients (mean age 3.9 years) completed all study phases. EUR-1008 was administered in 8 patients (mean age 3.5 years) by dispersion onto food, and in 11 patients (mean age 4.3 years) by capsule. For the entire group, 57.9% of patients were determined to be responders at the end of the study. Responder rates were similar for the 2 dosing options (62.5% for dispersion onto foods vs 54.5% for oral administration). EUR-1008 led to decreased stool frequency, and oil and grease in stools regardless of dosing option.

Conclusions: EUR-1008 was effective in children aged 1-6 years with CF and EPI. Exploratory analysis revealed that the effectiveness of EUR-1008 was maintained when dispersed onto foods, providing a convenient dosing option in very young patients and in older patients who have difficulty swallowing.

Wooldridge, J.L. 1 ; Heubi, J.E. 1 ; Lee, C. 2 1. Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA; 2. Eurand Pharmaceuticals, Inc., Vandalia, OH, USA Background: The majority of cystic fibrosis (CF) patients have exocrine pancreatic insufficiency (EPI), resulting in maldigestion and malabsorption of nutrients that requires supplementation with pancreatic enzyme products (PEPs). H2 antagonists or proton pump inhibitors (PPI) are often administered to improve the release of enzymes into the proximal intestine. In this pivotal study, we evaluated EUR-1008, a novel, entericcoated, zero-overfill PEP, without the concomitant use of gastric-acid modifiers.

Methods: This randomized, double-blind, placebo-controlled, crossover multicenter trial was conducted in patients aged ≥7 years with confirmed CF and EPI. The primary end point was change in coefficient of Overall our data demonstrate that indeed the defect in CFTR is linked with fatty acid abnormalities which seem to be independent of nutritional status. Additionally, cells from CF patients can respond to fenretinide treatment providing additional support that fenretinide should be considered as a potential therapy for cystic fibrosis patients. Background: Liver disease is often seen in cystic fibrosis (CF). Hepatic steatosis occurs in 20-73% of individuals with CF. Possible etiologies include malnutrition, essential fatty acid and choline deficiency or as yet undetermined dietary, environmental and genetic factors. Choline deficiency is associated with non-alcoholic steatohepatitis; supplementation with choline or its oxidative metabolite betaine, has improved or reversed hepatic steatosis.

Objective: To describe liver fat content and calf muscle fat and choline content using MRS in subjects with CF and pancreatic insufficiency (PI) ages 10-18 years at baseline and at 12 months.

Methods: Subjects were enrolled in an 18 month prospective doubleblind placebo-controlled supplementation trial with LYM-X-SORB, a novel lipid matrix containing choline and essential fatty acids. Height (cm) and weight (kg) were measured, BMI and Z scores were calculated, and liver enzymes (ALT, AST, GGT) were obtained on all subjects. Subjects underwent MRS scan (1.5T Siemens Avanto Whole Body Scanner) after a standard 11 hour fast. Muscle/liver Proton MRI and Spectroscopy were acquired with a Siemens Extremity coil/ body array, respectively. Localized waterunsuppressed spectra were acquired from similar volumes of interest (VOI, 18cc) selected in the liver using STEAM pulse sequence (TR=4 sec, TE=20ms, TM=10ms, spectral width=1KHz, 1024 complex data points, and 1 average). Chemical Shift Imaging (CSI, TR=1.7s, TE=30ms, spectral width=1KHz, and 4 averages) with and without water suppression was acquired from similar regions in the soleus muscle.

Results: Baseline measurements were obtained on 21 subjects (14 male, 7 female; 13.5±2.4 yrs),and 12 month measurements on 5 subjects (4 male, 1 female; 14.4±1.8 yrs). At baseline (in 21 subjects), variable liver fat (0.2 to 35%) and muscle choline content (0.04 to 1.03%) was noted, growth status was suboptimal, and liver enzymes were elevated (Mean ± SD: ALT, 49±20 U/L; AST, 50±36 U/L; GGT, 27±13 U/L). Liver fat and muscle choline levels were not associated with age, growth status, or liver enzymes. At 12 months, weight and BMI Z scores improved significantly in the 5 children with follow-up data (p<0.005); there was no significant change in liver enzymes, liver fat and muscle choline content.

Conclusion: Variable levels of liver fat and muscle choline are seen in children and adolescents with CF and PI. Preliminary data suggest improvement in growth status over one year, and no change in liver fat and muscle choline content or liver enzymes.

Supported by the NIDDK (R44DK060302), Clinical Translational Research Center, (UL1RR024134), Nutrition Center at Children's Hospital of Philadelphia and Avanti Polar Lipids Inc.

Data are expressed as mean±SD Colombo, C. 1 ; Crosignani, A. 2 ; Zazzeron, L. 1 ; Motta, V. 1 ; Battezzati, P. 2 1. Department of Pediatrics, CF Center, Fondazione IRCCS Ospedale Policlinico, Mangiagalli, Regina Elena, University of Milan, Milan, Italy; 2. Liver Unit, San Paolo Hospital Medical School, University of Milan, Milan, Italy To assess the long-term effects of ursodeoxycholic acid (UDCA) for the treatment of cystic fibrosis (CF) associated liver disease (CFLD) we performed cohort study including all the patients who were consecutively administered oral UDCA (20 mg/kg/day) at the CF Center of Milan between 1988 and 2007. Liver biochemistry, ultrasonographic and hepatobiliary scintigraphic parameters, nutritional status, pulmonary function and the occurrence of clinically relevant events were the study outcomes. During the period, 91 patients were treated with UDCA and evaluated for safety analysis; 59 were considered for efficacy analysis. Reasons for exclusion were absence of an adequate baseline evaluation (n=12), lack of compliance to drug administration or follow-up (n= 10), cirrhosis with portal hypertension at baseline (n=7), treatment duration lower than 12 months (n=2), concomitant diagnosis of biliary atresia (n=1). Routine visits were scheduled for all patients at least every six months. The effects of UDCA were evaluated in comparison with baseline values by using the 2-sided Wilcoxon Signed Rank Test at an α-level of 0.05.

Safety analysis (n=91): 7 deaths were observed, none due to CFLD. No adverse drug reactions were recorded. Efficacy analysis (n=59): mean UDCA treatment duration was 6.2 + 4.1 yrs, male to female ratio 38/11, history of meconium ileus was present in 38%; mean age at diagnosis of CF and CFLD were respectively 0.8 + 2.0 and 5.4 + 6.5 yrs; UDCA was started at 9.4 + 6.5 yrs of age and mean age at time of last observation during UDCA treatment was 15.6 + 7.3 yrs. Of these patients two died (not for CFLD) and 57 are still alive. Biochemical data are available for all patients after one year of treatment, for 51 after 2 years, for 34 after 5 years and for 12 after 10 years of treatment. A significant improvement was observed for ALT values after 1, 2 and 5 years of treatment compared with baseline (p< 0.05), whereas no statistical difference was observed in the small number of patients treated for 10 yrs. A significant improvement was observed only after 1 and 2 yrs of treatment for AST (p<0.05) and GGT (p< 0.05). Among hepatobiliary scintigraphic parameters, time of intestinal visualisation (TIV) significantly decreased from 24 + 26 to 20 + 23 min. after 1 year of treatment (n=32; p<0.05) and to 17 + 22 min. after 2 yrs (n=25; p<0.05). No significant difference was found during treatment for serum bilirubin concentrations, nutritional status and pulmonary function.

Conclusions: Long-term treatment with UDCA in CFLD patients was shown to be safe. The significant decrease of TIV is consistent with an improvement in bile secretion. Indices related to cholestasis and cytolysis decreased during the first 2-5 years, with no further improvements during subsequent follow-up thus suggesting that UDCA may slow but not halt the progression of CFALD. No significant effect of UDCA administration was found on other clinically relevant parameters such as nutritional status or pulmonary function. Introduction: Cystic fibrosis (CF) patients have chronic pancreatic insufficiency leading to malabsorption of fat soluble vitamins including vitamin D which may contribute to poor skeletal health, depression, carcinogenesis, type 1 and 2 diabetes mellitus and to muscular and respiratory dysfunction.

Objective: To assess the seasonal variability in the serum concentration of 25-hydroxyvitamin D ([25(OH)D]) and correlate with bone density, forced expiratory volume in 1 second, renal function, pancreatic insufficiency, class mutation, body mass index (BMI), diabetic status and gender. sis is elevated in the liver and other tissues (White et al. Am J Physiol Lung Cell Mol Physiol 2007;292:L476), but is not an energy-yielding substrate, indicating redistribution of lipogenic carbon away from the synthesis of energy-yielding substrates. These studies provide the groundwork to determine the cause of growth retardation in CF patients and mechanisms of regulation of metabolic fluxes in CF. for cystic fibrosis (CF) has increased the number of CF infants detected. For these infants, feeding advice is required and, if the infant is pancreatic insufficient (PI), pancreatic enzyme replacement therapy (PERT) prescribed. The choice of feeding regimen and of pancreatic enzyme preparation can be difficult. Although there is a body of evidence to support the use of breast feeding in non-CF infants, it is not universally accepted that this is sufficient for CF PI infants. If powdered enzyme preparations are used, this may lead to excoriation of mother's nipples and of the infant's lips and mouth resulting in early abandonment of breast feeding and change to use of infant formulas. Enteric coated (EC) enzyme preparations can prevent these problems, but there is little evidence to guide dosage in this age group and to confirm that breast-fed CF infants on EC PERT exhibit satisfactory nutrition and growth.

Aim: To evaluate results from breast or formula feeding in CF PI infants attending our CF clinic in Vancouver.

Method: Our clinic practice has been to prescribe EC PERT for all breast-or formula-fed CF PI infants. We support mother's choice whether the infant is breast-or formula-fed. All infants receive supplemental NaCl (2-4 mEq/kg/day). We have compared growth and nutritional parameters for 16 CF PI infants who were breast-fed until at least 6 months of age with 14 formula-fed CF PI infants. Diagnosis of CF was confirmed by sweat chloride analysis, and pancreatic status determined by measurement of fecal elastase using Schebo-Tech reagents.

Results: Initial PERT dosage after diagnosis was 5916 u lipase/kg for breast-fed infants, then increased to average 7082u lipase/kg at 6 months (range 2580-12009). For formula-fed infants, starting dose was 5419 u lipase/kg, then increased to average 6316 u lipase/kg (range 1341-11053 u lipase/kg).

Mean weight percentiles (CDC US Growth Charts 2001, Ross Pediatrics) for breast-fed infants was 17% at diagnosis, increasing to 28.5% at 6 months and 33.8% at 1 year. Formula-fed infants had mean weight percentile 24.4 at diagnosis, 35.9% at 6 months, and 32% at 1 year.

Mean albumin level for breast-fed infants was 24.6gm/L at diagnosis, increasing to 34 at 6 months and 40.8 at 1 year. Formula-fed infants had mean albumin of 27.5gm/L at diagnosis, 35.4 at 6 months, and 39.4 at 1 year (Normal albumin range for infants 6 days-1 yr. is 34-42 gm/l).

Conclusions: Use of EC PERT at per kg dosages similar to those used for older children allows breast feeding of CF infants. Breast-and formulafed infants showed equivalent increases in growth and nutritional parameters. Results were in the normal range 6 months after diagnosis and at 1 year of age.

The authors wish to thank Drs. D Peacock, and A Kielska for their valuable contributions to the care of these infants.

A Retrospective analysis over 4 yrs of 111 adult cystic fibrosis patients medical records attending the Cork Adult Cystic Fibrosis Centre was performed examining annual serum levels of the fat soluble vitamins including seasonal serum [25(OH)D]. The Mann Whitney U Test and Spearman correlation were used to correlate serum [25(OH)D] with left hip Z score, FEV1 % predicted, estimated Creatinine Clearance by the Cockroft and Gault Equation (eCrCl), age, pancreatic insufficiency, class mutation, BMI, diabetes status and gender.

Results: The overall prevalence of vitamin D deficiency [25(OH) D < 75nmol/l] was 86.5% in the winter months and 61.3% in summer months. Severe vitamin D deficiency [25(OH) D < 37.5nmol/l] was 53% in winter months and 17% in summer months. Summer and Fall serum vitamin D levels were found to be 72% and 69% higher than Winter and Spring levels with mean levels of 73.1nmol/l, 71.6 nmol/l, 42.4nmol/l and 43.01nmol/l respectively. In this study there was no significant association between mean serum [25(OH) D] and FEV1 per cent predicted after controlling for age, gender and BMI, nor did serum [25(OH) D] correlate with left hip Z score, pancreatic insufficiency, class mutation, diabetes status or renal function.

Conclusion: There is a significant seasonal variability in serum [25(OH) D] with a high prevalence of vitamin D deficiency in the Adult CF population despite standard vitamin D supplementation strategies. We suggest measuring winter and spring time serum [25(OH) D] as a more accurate guide to determining vitamin D insufficiency and that more aggressive vitamin D replacements strategies are needed in the adult CF population particularly at Northern Latitudes.

Bederman, I.; Freedman, J.; Drumm, M. Pediatrics, Case Western Reserve University, Cleveland, OH, USA Cystic fibrosis (CF) patients display short stature and low BMI, both strong predictors of pulmonary function and overall longevity. These phenotypes are largely attributed to malnutrition resulting from the early loss of exocrine pancreatic function, leading to nutrient malabsorbtion that affects growth and development. However, aggressive nutritional therapy, even at very early ages, does not fully restore the BMI of patients. Even with the best of nutritional care, patients have short stature and low BMI. This indicates that growth impairment is a multifaceted phenomenon in cystic fibrosis and brings into question the causative role of malnutrition.

Body mass is composed of fat and lean compartments that dictate growth rate and BMI. The changes of those compartments are controlled by the metabolic fluxes, i.e. rates of synthesis and degradation of lipids and proteins. CF mice (null and ∆F508) exhibit similar growth retardation as patients, but without pancreatic insufficiency or apparent malabsorption, providing a model to study the rates of metabolic fluxes that "set" the growth retardation phenotype. We found no difference in fatty acid absorption by comparing fatty acid profiles from feces of WT and CF -/mice relative to that of their chow. WT and CF profiles of fecal fatty acids looked virtually identical indicating similar absorption efficiencies. Using CT and MRI, we found that ~17% lower body weight in CF mice was composed of 17% less of whole-body adipose tissue and 7% less of lean muscle tissue. In CF mice, epididymal adipose tissue was reduced by a third (598±79 vs. 399±33 mg, WT and CF, respectively, p<0.05). We hypothesized that such drastic differences could be due to excessive lipolysis and/or decreased synthesis rate. We found free plasma glycerol (a measure of intracellular lipolysis) was indeed significantly elevated (0.14±0.01 vs. 0.23±0.02 mM, WT and CF mice, respectively, p<0.01). To determine fractional rates of synthesis of newly made lipids and proteins, we injected mice with a bolus of deuterated water ( 2 H 2 0) and determined 2 H incorporation into newly made fatty acids and proteins of adipose tissue, skeletal muscle and liver. We found fractional lipid synthesis was not different in the adipose tissue (14.0±1 vs. 13.2±0.8%, WT and CF mice, respectively), but absolute rate of synthesis was significantly lower in CF (85.3±15.7 vs. 40.2±6.4 mg/day, p<0.05). Fractional rate of protein synthesis tended to be lower in both skeletal muscle and heart proteins. However, absolute rate of whole-body protein synthesis was significantly lower in CF mice (~45%, p<0.05). Lastly, we found significantly decreased hepatic de novo lipogenesis in CF mice (~24 and ~20%, palmitic and stearic fatty acids, respectively, p<0.05) while cholesterol synthesis was significantly increased in CF mice (~20%, p<0.05).

These findings indicate that decreased size in CF may involve a combination of altered rates of protein and lipid biosynthesis. Cholesterol synthe-

Keller, B.O. 1 ; Bay, B. 2 ; Innis, S.M. 2 1. Pathology and Laboratory Medicine, University of British Columbia, Vancouver, BC, Canada; 2. Pediatrics, University of British Columbia, Vancouver, BC, Canada Liquid chromatography combined with mass spectrometry (LC-MS) allows simultaneous detection of hundreds of components in biological samples, such as urine, plasma and tissue extracts. Subsequent data processing using sophisticated multivariate statistical analytical software enables reliable, relative quantitative comparison of large numbers of metabolite levels between sample cohorts. This allows broad ranging analysis seeking to identify disturbances in metabolic pathways that may be causally linked to or predict subsequent disease. Our focus is the biochemical basis for disturbed methyl and thiol metabolism and its relation to varying degrees of disease severity among children with cystic fibrosis (CF). Using this approach, we analyzed 40 urine samples from children with CF and 9 healthy children without CF. All of the children with CF were outpatients at the CF clinic of the British Columbia Children's Hospital and all were seen during a routine outpatient clinic appointment. Among metabolite differences, hippurate levels were consistently lower in urine from children with CF compared to children without CF.

Hippuric acid is the hepatic conjugation product of benzoic acid. After activation to its coenzyme A ester and conjugation to glycine, it is eliminated via renal excretion (1) . Benzoic acid is a natural component of foods, such as berries and other fruits and, importantly, is widely used in high doses as a preservative in processed foods and soft drinks. Due to widespread use of benzoic acid in foods and no evidence of lower benzoate intakes among children with CF when compared to other children, metabolic limitation of conjugation appears likely. Gregus et al. (2) have shown that the limiting factors for benzoate metabolism are glycine and/or coenzyme A, not low enzymatic activity. Increased glycine loss via bile acid conjugates may deplete glycine pools; and glycine is also an essential amino acid in glutathione and involved in methyl transfer, all known to be disrupted in CF (3) , but Coenzyme A is also formed from sulphur amino acids. Intakes of benzoate that exceed glycine availability lead to increasing benzoate levels in plasma, since renal excretion of unconjugated benzoic acid is minimal and other benzoate metabolic pathways, such as formation of benzoylalanine or benzoylglucuronide, represent very limited alternatives (1) . Current analysis is focusing on quantitative analysis of plasma benzoate to confirm impaired conjugation and excretion, also enabling assessment of potential risk of benzoate toxicity, as well as quantitative measures of methyl metabolites and amino acids.

Supported by a grant from the CFF. We thank Roger A. Dyer for his help with sample analysis and data collection. Background: Cystic fibrosis related diabetes (CFRD) is the most common co-morbidity in persons with CF. Several studies have provided evidence for the clinical utility of oral glucose tolerance testing (OGTT) screening in adolescents and adults with CF. Guidelines recommend annual OGTT screening for all CF patients age 10 and older. This recommendation does not address younger patients with CF. However, OGTT screening in children could be recommended if, as in adults, the glucose tolerance category provided clinically relevant prognostic information. This current retrospective chart review was performed to determine whether the baseline OGTT category in CF children age 6-10 years predicts subsequent clinical course.

Methods: Data were retrospectively retrieved from the Minnesota CF Center Database on all patients who had an OGTT performed between Jan 1, 1998 and Dec 31, 2003 , to identify individuals between the ages of 6-10 years. Children were classified as normal glucose tolerance (NGT) or abnormal glucose tolerance (AGT) based on OGTT. Abnormal glucose tolerance was defined as impaired glucose tolerance (IGT), which is a 60 min glucose >140 mg/dl and <200mg/dl, or indeterminate glucose tolerance (IND), which was a glucose level >200 mg/dL during the test, but a glucose level <140 mg/dl at 60 minutes. Height, weight, FEV1, FVC and hemoglobin A1c (HbA1c) were collected at baseline and 5 years later. Children without 5 years follow-up data were excluded. All patients and their parents gave informed assent/consent for their records to be reviewed.

Background: There are reports of an increased prevalence of hypogonadism in the adult male cystic fibrosis (CF) population, but the limited published data on this issue has produced conflicting results. Information on predictors of hypogonadism in the CF population is also lacking.

Aim: Determine whether adult men with CF have a higher prevalence of hypogonadism based on total serum testosterone levels when compared to the general population and to determine predictors of hypogonadism that exist in this population.

Method: A retrospective chart review was completed for all men >18 years of age seen at least once at the University of Minnesota CF center between January 1, 2008 and April 30, 2009. Results: A total of 165 adult men were included, of which 118 (72%) had testosterone levels and/or were on testosterone replacement. Of these 118 patients, 90 had a DEXA scan obtained. Characteristics of the 118 patients are summarized in the table below.

There was no significant difference in age, BMI, PFT, or DEXA results between those with and without testosterone levels. Patients with testosterone levels were more likely to have cystic fibrosis related diabetes (55% vs 21%, p-value of <0.001) than those without levels. Logistic regression analysis found age, BMI, PFT, diabetes status, and bone mineral density were not predictors of hypogonadism (testosterone level < 10nml/L).

Conclusion: These results suggest adult men with CF have an increased prevalence of hypogonadism. We did not identify predictors of hypogonadism. Further study is needed to determine the clinical relevance of these findings. Results: Of 43 children who had at least one abnormal OGTT between 1998 OGTT between -2008 had sufficient data to be included. Thirty-five children with CF and NGT matched for age and gender were included as controls. At baseline, no children in either group had fasting hyperglycemia or impaired fasting glucose. There was no significant difference in height, weight, BMI or FEV1 at baseline between AGT and NGT. Diabetes subsequently developed in 14 children (40%) in the abnormal glucose tolerance group and in one child (3%) in the control group (p = 0.0024, Log-rank test from Kaplan-Meier curve for time to diabetes). The average age of onset was 13 +/-1 years in boys and 11+/-1 years in girls, considerably earlier than the average diabetes onset of 23 years in the UM CF population. At 5 year followup there was no significant difference in height, weight, BMI or FEV1 between AGT and NGT.

Conclusions: Children with abnormal glucose tolerance are clearly at high risk for progression to early onset frank diabetes compared to children with NGT. OGTT screening identifies these high-risk children. The understanding of the Cystic Fibrosis Related Diabetes (CFRD) pathogenesis and the prediction of its development are still challenging tasks. Starting from year 2003 patients who received routinely OGTT at the regional cystic fibrosis center of Milan were followed prospectively with major endpoints considered being the development of stable CFRD and death. The OGTT at recruitment was analysed for glucose tolerance and insulin secretory (beta cell glucose sensitivity) and insulin sensitivity parameters derived from modelling of glucose, insulin and c-peptide profiles.

Two hundred patients aged 11-35 yr were followed for 733 patient-years (median length of follow-up 4 years). During this time 13 subjects developed CFRD and 11 died with a annual incidence rate of CFRD of 1.8% and an incidence rate of deaths of 1.5%.

According to life-table analysis, cumulative incidence of diabetes at 1 and 5 years was 10±4% and 17±5% in the lowest tertile of beta-cell responsiveness compared to 2±1 and 4±3% in the upper tertiles (p=0.002 log-rank test), with an incidence rate of 41 and 7 cases/1000 patient-years, respectively; it was 8±4 and 17±6% in the lowest tertile of insulin sensitivity compared to 3±2% and 3±2% (p=0.025), with an incidence rate of 35 and 9 cases/1000 patient-years; it was 5±3 and 14±5% in the lowest tertile of FEV1% compared to 3±2% and 3±2% (p=0.05 log-rank test), with an incidence rate of 30 and 11 cases/1000 patient-years.

When glucose tolerance was considered, it appeared that diabetic tolerance involved a cumulative incidence of diabetes at 1 and 5 years of 27±11% and 41±13% compared to 2±1 and 5±2% (p=0.001 log-rank test), with an incidence rate of 107 and 11 cases/1000 patient-years. However, considering that OGTT presently is the gold standard for diabetes diagnosis, but less than half of the 15 patients initially diabetic at this test developed stable CFRD, and only 6 of 14 subjects who developed CFRD had shown a diabetic tolerance, the prediction of CFRD should also include the evaluation of insulin secretory and sensitivity defects as well as pulmonary function measurements. 5%, and 23% respectively before the intervention. Adherence to the CFRD protocol for treatment increased from approximately 30% at the start of CFRD clinic to 80% at present. The patient survey was completed by 41 of the 70 patients. The majority of patients were satisfied with their CFRD clinic experience, including understanding of referral, time of appointment, coordination of appointment with CF clinic, and frequency of appointments. However, 26% reported that CFRD clinic was not helpful and 32% reported that the visit did not improve their understanding of CFRD. Most (64%) of those dissatisfied were not prescribed CFRD therapy, and comments indicated discontent directed toward an additional diagnosis rather than dissatisfaction with the CFRD clinic itself. Regardless of CFRD therapy, most patients reported that the treatment for CFRD was easy to manage with reports of increased difficulty increasing with complexity of treatment. Insulin was perceived as slightly more complex as 25% of patients on oral medications reported that care was somewhat difficult vs. 37% of patients on insulin.

Conclusion Our experience suggests that the a dedicated CFRD clinic staffed by a single ES increases consistency of care and allows for the development of a therapeutic relationship with an experienced CFRD ES. This may decrease stress related to this additional diagnosis and may explain the low rate of difficulty with CFRD reported by our patients. In addition, patients report little difficulty on daily management of CFRD, achieving glucose control within recommended levels. Additional data is required to determine whether this intervention affects long-term outcomes. The median survival in cystic fibrosis (CF) is now 37.5 years. The Toronto Adult CF clinic follows 400 adults with CF with >20% being older than 40 years. With an aging CF population we anticipate the progression of hormonal changes leading to menopause. The purpose of this study is to gather information on the timing of perimenopause in women with CF and to characterize the symptomatology during this stage.

Methods: Women with CF age >30yrs were invited to participate in the study. Exclusion criteria were pregnancy, surgical menopause and post lung transplantation. A questionnaire was designed by the principal investigator (AT) with valuable input from the Canadian CF Nurses' Interest Group. The questionnaire consisted of demographic and CF related information obtained from the medical record and menstrual history and the Menopause Rating Scale (MRS) to capture symptoms of perimenopause completed by the participants.

Results: To date, 33 of 65 eligible women have completed the survey. These women have a mean age of 42yrs (range 31-63) with BMI 22 (17-35); FEV1 56% predicted (23%-115%) and 61% are PI. Eleven of 33 (33%) subjects answered questions in the MRS indicating symptoms of perimenopause. Five of 11 had reached menopause at an average age of 47.8yrs (45-51) and 6/11 had not yet reached menopause. The average age of onset of perimenopause symptoms was 37.8yrs (32-42yrs). Perimenopausal symptoms included: hot flashes in 58.1%, palpitations 46%, sleep problems 75%, depressed mood 68%, anxiety 59%, exhaustion 67%, sexual problems 64%, bladder problems 51%, vaginal dryness 53% and joint/muscle discomfort in 75%. Sleep problems had the highest score in severity (15.6% marked "very severe"). All but one patient reported increased respiratory symptoms which included: dyspnea 9/11, chest tightness/wheezing 8/11, thicker sputum 10/11, increased sputum volume 9/11, and hemoptysis 5/11. Conclusion: Women with CF appear to experience perimenopause sympotms slightly earlier than the general population (37.8yrs in CF vs 41yrs). Sleep problem is the most dominant and severe symptom reported. It is not clear if the increased respiratory symptoms are related to progressive lung disease or perimenopause. Educational material can be developed to increase patients' knowledge of perimenopause and outline strategies that can be used to deal with this change of life. Leptin, a protein secreted from white adipose tissue, plays an important role in the regulation of food intake. Serum leptin levels have been shown to be regulated by circulating insulin concentrations. Cystic fibrosis (CF) is often associated with endocrine pancreatic dysfunction. As a result, most CF patients have decreased and/or abnormal insulin secretion. Our hypothesis was that serum leptin concentrations would be reduced in CF subjects and particularly in those that have CF related diabetes (CFRD) who showed a pronounced decrease in insulin secretion. To examine the relationship between leptin and insulin, we measured the levels of both hormones in the fasting state in 10 healthy controls (6 men and 4 women), 50 CF adults with normal glucose tolerance (NGT, 30 men and 20 women), 22 CF subjects with impaired glucose tolerance (IGT, 12 men and 10 women) and 12 subjects with CFRD but without insulin treatment (6 men and 6 women). We also determined body mass index (BMI) and the percentage of fat mass by bioimpedance for each participant. We observed no difference in serum leptin levels between healthy controls and CF subjects from all three glucose tolerance groups (NGT, IGT and CFRD). In CF subjects we observed positive correlations between leptin and BMI (p<0.001), percentage of fat mass (p<0.001), fasting insulin (p ≤0.035) and insulin resistance as evaluated with the homeostasis model assessment (HOMA) index (p≤0.034). Interestingly, after correction for BMI and fat mass, the correlations between leptin and fasting insulin levels and HOMA index remained significant for women but not for men. Thus, while leptin levels of CF subjects are similar to those of healthy controls, we observed significant differences in the association between leptin and insulin among men and women. These results suggest that the association between these two hormones may be modulated by gender. Future studies are necessary to examine these differences observed between men and women. Background: Evidence indicates that the diagnosis of cystic fibrosis related diabetes (CFRD) may follow a period of decline in lung function and weight gain. Early diagnosis and treatment of CFRD may alter the progression of disease. In 2007 our center began annual oral glucose tolerance tests (OGTT) for patients with CF age ≥ 10 years. Inconsistency in the diagnosis and treatment of CFRD was noted among various endocrinology specialists (ES). Later that year, we evaluated current practice with a literature review, consensus meetings, and a benchmarking visit to a leading CFRD-CF center. As a result, we made three changes: a protocol for the diagnosis and treatment of CFRD; a CFRD clinic on the same day as CF clinic; and a primary ES for those with CFRD.

Methods: We assessed the effect of these changes by outcome data (BMI %, FEV1%, blood glucose (BG), hemoglobin (Hgb) A1c, CFRD treatment) and a patient satisfaction survey for the CFRD clinic and burden of care related to CFRD.

Results: Seventy patients were seen in CFRD clinic. Patients with treatment for CFRD and those with BG > 126 mg/dl and/or an insulin level > 53 IU/ml were referred to the clinic. This group of patients had stability in BMI%, FEV1%, BG, and Hgb A1c with medians of 48%, 80%, 120 mg/dl, and 5.6% respectively over a period of 16 months. Fifty-four percent were not prescribed any medication for CFRD, 14% were prescribed an oral hypoglycemic agent, and 32% were prescribed insulin, compared with 72%, recorded as a percentage of the patients' best FEV1 observed over the preceding year. The number of inpatient days and total number of days of IV treatment was recorded and the percentage of time spent as an inpatient calculated. Data were analysed using a Spearman's correlation (the data were not normally distributed).

Results: The mean percentage time spent as an inpatient was 63.3% equating to 9.1 days. FEV1 mean change was 18.2% equating to 0.54 litres. There was a significant although weak positive correlation between percentage time spent in hospital and percentage change in FEV1 (r=0.02, 95%CI 0.02 to 0.37, p<0.05).

Discussion: These results are consistent with Thornton et al. who concluded that hospital treatment was more effective at improving FEV1 than home treatment. This is perhaps not surprising due to the availability of regular supervised physiotherapy and assistance in IV preparation and administration as well as help with other activities of daily living. It should be noted that some patients prefer home treatment as it is less disruptive to their lifestyle as described by Balaguer and González del Dios. Whereas some patients are required to administer home IV antibiotics due to bed shortages it would be reasonable to identify those most likely to benefit from an inpatient stay. Further comparison after the introduction of a home care physiotherapy service may improve outpatient treatment outcomes.

Conclusions: Cystic fibrosis patients improve more in terms of FEV1 when treated in hospital with IV antibiotics than at home. This warrants further investigations taking into account both economic and quality of life issues.

Allen, S.L. 1 ; Ciolino, J. 2 ; Flume, P.A. 3 Background: Medical residents play an important role in the care of patients with cystic fibrosis (CF) in the hospital setting. We hypothesize that if medical residents have greater satisfaction with the inpatient CF encounter, that they are more likely to provide higher quality care for these patients. As part of a quality improvement initiative (QI) on inpatient CF care, we sought to assess medical resident satisfaction with various aspects of our CF inpatient program including education, communication and ancillary staff support; as well as attitudes toward CF patients themselves.

Methods: Of the 96 internal medicine residents at our academic medical center, 63 (66%) completed our QI survey. The 56-item survey consisted of questions regarding preparedness in various aspects of CF-related disease, general attitudes toward CF patients, and satisfaction with CF patient care, support, and education. Responses were based on 5-point ordinal scales. Internal validity was demonstrated. Statistical analyses consisted of Mantel-Haenszel chi-squared tests, Spearman's rank correlation coefficient calculations, and exact chi-squared tests, as appropriate.

Results: A positive correlation existed between year of training and comfort with preparedness with respect to managing nutrition, pain and anxiety (p<0.001). Greater satisfaction with support from nursing (p=0.04), respiratory therapy (p<0.01), and pharmacy (p=0.001) also correlated with year of training. The mean rating of overall satisfaction with care of CF patients was 3.02 (±0.84). Those who reported a negative attitude toward CF patients, as evidenced by a low rate of satisfaction, were more likely to describe them as a burden to the service (p<0.01) and reported there was little to learn from their care (p=0.001). A positive attitude toward CF patients did not correlate with year of training, but there was a positive correlation with satisfaction with bedside teaching (p<0.01), communication with the CF team (p<0.01), and level of autonomy (p<0.01).

Conclusion: As expected, comfort with managing inpatient CF care increased with level of experience but this did not translate into greater satisfaction with care of CF patients. It appears that increasing bedside education and improving communication between the CF team and the medical residents would improve overall satisfaction of residents with regards to the CF patients. This may, in turn, improve patient satisfaction and quality of care, and provides us with a focus for a QI project. regarding high-calorie foods to assist them in meeting their calorie goals. Therefore, we devised a quality improvement project that would provide patients with a schedule for meals and snacks, as well as education on highcalorie food choices available in the hospital in order to improve their calorie intake and weight gain.

Methods: As part of an improvement project led by a multidisciplinary steering committee (that included parents) whose intent was to standardize hospital treatment of pulmonary exacerbations and make them more efficient and productive, we implemented a daily schedule that allocated patients adequate time for three meals and several snacks to better meet their calorie goals. We also provided education on high-calorie meal and snack choices in the hospital and asked them to keep food records during the hospitalization, to encourage patients to actively participate in reaching their daily calorie goals. A token incentive program was introduced to promote compliance with these tasks. For those patients that complied with the program, we calculated their estimated calorie needs based on the Dietary Reference Intakes multiplied by a factor for weight gain and compared their intake with that estimated need. We reviewed the records of patients with suboptimal nutritional status (a BMI less than the 50th percentile) and compared the weight gain of those who participated in the pilot program with the weight gain of those who did not during the course of the hospitalization.

Results: Since beginning the nutrition pilot program in March 2009, a total of 10 patients with a BMI less than the 50th percentile were enrolled. Of that group, 6 patients (60%) filled out food records for an average of 2.4 days during their admission. Those patients' intake was 108.24% of their estimated calorie needs. We also compared 15 patients with a BMI less than the 50th percentile admitted during this same time period, the 10 patients who participated in the pilot program with 5 other patients who did not participate. Those that participated averaged a 4.01% weight increase during the hospitalization, and those who did participate averaged a 1.53% weight increase (p<0.36, unpaired T-test).

Conclusion: While this data is not statistically significant due to small sample size, we believe they show a trend toward increased weight gain in patients who receive a daily schedule to allow adequate time for meals and snacks, as well as nutrition education on high-calorie foods. Based on the preliminary pilot data, we plan to continue this program and extend it to all hospital admissions. Background: Cystic fibrosis (CF) is a progressive disease that is characterised by repeated respiratory infective exacerbations, a gradual decline in lung function and eventually respiratory failure in >90% of patients. Prolonged treatment with high dose intravenous (IV) antibiotics is an established treatment for exacerbations which can be delivered in hospital or at home. Thornton et al. (J Cyst Fibros 2005; 4:239) concluded that hospital treatment was significantly more effective at improving FEV1 but more expensive than home treatment. In a review of the literature Balaguer and González del Dios (Cochrane Database Syst Rev 2000;(3):CD001917) identified one randomised controlled trial that included 17 CF patients. The authors suggested that home IV treatment in CF showed no differences for clinical outcomes, adverse events, or complications. Patients who did the home IV regimen however, reported being more tired and found the treatment more difficult to master. A possible reason for these findings was cited as patients being more active and needing more support. Home therapy was cheaper for families and the hospital.

Aims: The aim of this study was to retrospectively evaluate the effectiveness of home versus inpatient IV antibiotics on lung function in patients attending the All Wales Adult CF Centre between August 2008 and Febuary 2009.

Methods: A retrospective analysis of 123 courses of IV antibiotics in 89 patients was conducted. The change in FEV1 from pre-to post IVs was The Cystic Fibrosis Foundation (CFF) Therapeutics Development Network (TDN) facilitates the clinical testing of new therapeutic agents for cystic fibrosis (CF). The CF Research Coordinator Mentoring Program (CFR-CMP) was developed to provide educational resources and a networking environment for CF research coordinators to promote quality improvement in clinical research. CFRCMP is modeled after a program designed by CF dieticians to promote quality improvement in clinical care. Roles include: Team Leader (leads the overall effort of the program), Facilitator (collaborates with the Team Leader and oversees mentor/apprentice pairs), Mentor (experienced research coordinator), Apprentice (research coordinator either new to CF or to clinical research), and the Executive Committee (1 member each from CFF/TDN-CC, the Team Leader, 2 Facilitators/1 Mentor.) The Committee is responsible for program oversight and approval of matches.

The CFRCMP is being rolled out in phases. Phase 1 of the program (2008-2009) is currently underway and is designed to improve site clinical research programs by establishing a forum for RC education and professional growth. The initial set of mentor/apprentice pairings was selected by the CFRCMP Executive Committee. Consideration was given to site location/size, types of studies conducted at site, potential mentor and apprentice experience/education, and additional site support. The apprentice was required to visit the mentor site for a 1-2 day site visit to observe best practices. Site visits were completed by November 2008.

Evaluations provided to mentors (n=10) and apprentices (n=10) were completed by 95% of participants, all respondents either agreeing or strongly agreeing that the program met expectations. Nearly 90% of respondents felt properly prepared for site visits. The majority of apprentices (80%) felt that participation improved the quality of research at their center, while 90% felt that they had increased knowledge of CF clinical research.

The accomplishments of Phase 1 include development of an online application, mentoring information sheet, Clinical Research 101 document, and RC webpage on https://CFCLINICALRESEARCHNET.CFF.org for centralized networking. Several lessons were learned from Phase I: leadership is integral to success; identification of measurable objectives is key; appropriately tracking metrics are essential; and communication among team members is pivotal to quality improvement.

For Phase 2 (2009), additional information will be requested from applicants. Principal investigators will be better integrated into the program. New tools include agenda template, objectives development worksheet and site visit report. Support provided by the Cystic Fibrosis Foundation. Background: A number of clinical scores have been developed to monitor health decline in patients with cystic fibrosis (CF). One of the more comprehensive scores is the "predictive 5-year survivorship model" developed by Liou et al.(2001) . This "Liou score" includes demographic variables, pulmonary infections, co-morbidities, and measures of pulmonary health. Although initially designed to predict 5-year survival, decline in this score over time may be helpful to identify patients that may need more aggressive intervention. This abstract describes the mean decline in Liou audited. Seventeen were either seen in the office within 24-48 hours (5), admitted (3), or received oral antibiotics over the phone (9). Of those 17 patients, 13 (76%) had follow-up appointments made and attended.

Conclusion: A modified Phone PES identifies pulmonary exacerbations over the phone, leads to rapid initiation of antibiotic treatment, and ensures timely office follow-up. The follow-up rate remains sub-optimal, so future audits will investigate why some patients are not being seen after antibiotics called in over the phone are completed. In addition, future audits will investigate how many patients who call frequently are being seen every three months, as recommended by the CF Foundation.

Swartz, J.L.; Eddy, J.C. Vermont Childrens Hospital, FAHC, Burlington, VT, USA Background: The Vermont Cystic Fibrosis Program is part of the Children's Specialty Center, an outpatient subspecialty clinic area composed of 19+ pediatric subspecialty practices. We hold weekly CF clinics, have monthly team meetings which include members of the family advisory board, and attend the monthly CF advisory board meetings. As part of transitioning our patients to the adult program, one of our CF Team goals is to develop educational tools related specifically to the CF patient. We felt that the team members routinely educated the families, have addressed education related to transition for the 12-18 year old, but did not have a formal process in place to educate the patient from diagnosis to the age of transition. As most patient teaching was being done by the CF Nurse Coordinator, we were limited by time and priority of other tasks related to a busy clinic. Two years ago, the Children's Speciality Center applied for and received funding from the Children's Miracle Network to support a full time Child Life Specialist (CCLS) for our oupatient subspecialty practices. At our institution, CCLS's were primarily staffed in the inpatient setting, so their role had not been defined in the outpatient areas. The Child Life Council defines the role as "Child life specialists are experts in child development, who promote effective coping through play, preparation, education, and self expression activities."

Our CF Team was able to describe a potential role of patient education through medical play.

Method: During the first year, the CCLS attended monthly staff meetings, pre and post clinic conference, shadowed each of the team members, and met with the Nurse Coordinator to further define her role. The focus of the first year was to establish a relationship with the CF patient and family by providing diversional activities, evaluating compliance and creating compliance take home charts, introducing medical play related to procedures, and accompanying patients to radiology and the laboratory. She also provided diversional activities during patient sweat tests. With the CF Team and a parent from the family advisory, she was able to create a Patient Checklist which identified each member of the team by name and photo, each procedure by illustration, and a game to be played. Each assigned discipline and procedure specific to that patient was highlighted by the CCLS. This was the patient's tool to use, checking off each area when done. This provided the patient with a sense of time and knowledge of the CF clinic, and when it was complete.

The goals for the second year were to develop educational tools related to each procedure (HRCT, bloodwork port flushes, etc), and to develop age specific skill sets related to each discipline using printed and hands on materials, games and computer access to CF Voice and Starbright. (A nutrition skill set has been completed.)

Future plans include expansion of skill sets to include respiratory, nursing and social work, and review and expansion of skill sets for ages 12-18 years.

Conclusion: The addition of a Child Life Specialist to our CF Team has allowed for the creation of developmentally appropriate educational tools, increased comfort with medical procedures, and increased time for the rest of the team members to spend with the family. score in a pediatric CF population, and the potential use of this tool in clinical practice.

Methods: At our CF center a multidisciplinary, comprehensive "case review" of each pediatric patient occurs every other year. It includes discussion of the patients' chronic therapeutic interventions, airway clearance, development, nutritional status, genetics, research study eligibility, clinic attendance, pulmonary function, pulmonary exacerbations with courses of treatment, and changes to their psychosocial situation since the previous case review. Beginning in Feb. 2007 Liou scores were calculated for all pediatric patients seen at the CF center at Akron Children's Hospital as part of this comprehensive review. Scores were calculated every two years. All patients with at least two scores were included in the analysis. Mean decline in Liou score was determined for males and females and the 6-12 y, 13-18 y and 6-18 y age groups.

Results: Fifty-four patients 6-18 y had two or more Liou scores calculated. The mean decline in score for the population over 2 years was 12.1+/-29.4. For patients 6-12 y the decline was 6.2+/-32.1 and for 13-18 year old patients 13.9+/-28.6. The mean decline for male patients was 14.4+/-29.3, and for female patients, 9.0+/-29.8. There were no statistically significant differences between groups. The majority of patients (66.7%) had a decline in score over 2 years (mean -26.4+/-24.1; 39% female), and 18 patients had an increase in score (mean + 16.5+/-13.8; 50% female).

Conclusions: Liou score trends provide an additional objective measure of a CF patient's disease progression. A decline in score may be an indication to re-evaluate current treatment regimens and to discuss available chronic therapies and/or aggressive nutritional support. The benefit of adding therapy at the time of a drop in Liou score may outweigh the perceived negative aspects of adding additional time-consuming and costly treatments. A stable or increased score may indicate that therapies and treatments in place are appropriate. In addition, Liou score trends may be a valuable tool that could be used to improve the prescription and use of recommended, chronic CF therapies. We describe the value of using a standard scoring system to identify pulmonary exacerbations via telephone triage and to ensure subsequent office follow-up. A telephone triage form was adopted in 2006 to rapidly identify and treat pulmonary exacerbations using a modified version of the Pulmonary Exacerbation Score (PES) that has been in use in our outpatient clinic as a result of the Learning and Leadership Collaborative project in 2004. This phone-adapted modification of the PES assists nurses to consistently assess pulmonary exacerbations over the phone so that antibiotic treatment can be initiated quickly, and office follow-up assured.

Method: The Phone PES form includes the items from the PES broken down into two categories: Systemic Symptoms/Signs (fever, malaise or fatigue, absenteeism from work/school and anorexia/poor appetite in the last 2 weeks), and the Pulmonary Symptoms/Signs (increased cough frequency, duration or intensity, major changes in sputum or chest congestion, increased shortness of breath and presence of hemoptysis). All symptoms/signs are assigned a numeric score which are added together to give a modified Phone PES, indicating if further evaluation and/or treatment is warranted. A Phone PES ≥3 indicates high potential for a pulmonary exacerbation, and these patients by our protocol must either receive treatment or have an appointment scheduled within 24 to 48 hours. In addition, nurses record the date of the last clinic visit and recent antibiotic use (oral or intravenous) on the form. If the patient has not been seen in the last 3 months, or if they received antibiotics since the last appointment, they are also scheduled to be seen within 24-48 hours.

Results: Since implementation in 2006, two audits have been conducted on the use of the phone PES and the success of initiating treatment and assuring optimal follow-up in the clinic. During the first audit (in 2007) 19 charts were audited. Ten had a PES ≥3. Nine were either seen in the office within 24-48 (1), admitted (0), or received oral antibiotics over the phone (8). Of those 9 patients, 8 (88%) had a follow-up appointment made and attended that appointment. In 2009, 20 charts with a phone PES ≥3 were

Validity of a modified shuttle test in adult cystic fibrosis

The endurance shuttle walk: A new field test for assessment of endurance capacity in COPD

CAN THE PERFORMANCE OF TESTS FOR NEONATAL SCREENING FOR CF BE IMPROVED? Dankert-Roelse

ACID SUPPRESSION THERAPY AS A RISK FACTOR IN CLOSTRIDIUM DIFFICILE INFECTIONS IN PEDIATRIC CF PATIENTS

1,2 1. Pharmacy, Intermountain Primary Children's Medical Center

Cystic Fibrosis department

Basil Elnazir (2) and Peter Greally (2)

Department of Clinical Nutrition and Dietetics

We conducted a study in which our aim was to assess plasma zinc levels and to determine if any relationship exists between zinc levels and albumin, vitamin A, height, BMI, and FEV 1 in children with CF. Method: A retrospective review of plasma zinc levels from annual review in children with CF was conducted. Blood samples for zinc analysis were obtained, pre-breakfast in a fasting state. Albumin, vitamin A, height, BMI, and FEV1 were also recorded. BMI and height measurements were converted to Z scores to allow for more accurate comparison. Mintab statistical package was used to analyse the data. Results: Forty patients (16 girls), mean age of 11.9± 4.6 (SD) were included in study. Five patients were pancreatic sufficient (PS)

Dietary Reference Intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium and zinc

Liver stiffness was more than 8.8 kPa (value associated with F2 Metavir category in chronic viral hepatitis) in 15 children. Score was over 14.6 kPa (value associated with cirrhosis) in 5 cases. Liver stiffness was correlated with portal hypertension (p<0.001); liver stiffness was also correlated with APRI

VITAMIN D SUPPLEMENTATION IS ASSOCIATED WITH IMPROVED SERUM 25-HYDROXY-VITAMIN D LEVELS IN CHILDREN WITH CF

1,2 ; Zhang

Vitamin D insufficiency is common in children with cystic fibrosis (CF). The 2004 Cystic Fibrosis Foundation (CFF) Consensus Report on bone health recommended vitamin D supplementation to maintain serum 25-hydroxy-vitamin D [25(OH)D] at 30-60 ng/ml. However, recent studies showed that the CFF supplementation regimen was ineffective in improving serum

Methods: A total of 131 patients (age 0.8 to 22 years) who had serum 25(OH) level documented in medical records between 1/2006 and 10/2008 (about 18 months before and after implementing the vitamin D supplementation protocol) were studied. Treatment for vitamin D insufficiency is based on serum 25(OH)D level (mild: 20-29 ng/ml; moderate: 10-19; severe: <10). For children weighing <40 kg, 50, 75 or 100 IU/kg of vitamin D (cholecalciferol preferred over ergocalciferol) are prescribed for mild, moderate, and severe vitamin D insufficiency, respectively. For children weighing ≥40 kg, 3000, 4000, or 5000 IU/day are prescribed according to level of insufficiency listed above. Follow-up serum 25(OH)D is obtained after 3-6 months of supplementation

Multivariate analysis revealed vit D supp was the only significant modifier of this relationship (F=3.99,p=0.047). The strongest association between PFT and vit D existed for 34 (17%) subjects taking no vit D supp, both for FEV1 (r=0.61, p<0.0001) and FVC (r=0.53, p=0.0013). These 34 patients did not differ from the entire group in any of the variables tested. No significant relationship existed between PFT and vit D in patients taking any vit D supp. Vit D supp did not correlate with serum vit D levels, regardless of dose. Conclusions: Serum 25-hydroxy vitamin D levels are significantly associated with pulmonary function in CF, particularly in older patients and seasons with greater sunlight exposure. The lack of an association in sub

Age range was 1-65 years old. Most cases of CFRD in patients with cirrhosis were found in patients over 20 years old. Data Analysis: A chi-square test using SAS software was constructed. It demonstrated a significant difference in rate of CFRD in patients with cirrhosis, compared to CF patients without cirrhosis (p<.0001). The odds ratio was 4.788 (95% CI 2.49, 9.17). Conclusions: CFRD is more common in CF patients with cirrhosis compared to the overall CF population (p<.0001). Patients with CF and cirrhosis have 4.8 times the odds of having CFRD. This may reflect a common pathogenic pathway. The presence of both complications presents a challenge in managing these patients. Our data

Cftr-/-) have reduced bone density. Given these two findings, we hypothesized that Cftr heterozygote mice (Cftr+/-) will demonstrate a partial bone density phenotype and increased markers of bone resorption. To test this notion

Bone mineral density was measured by peripheral quantitative computerized tomography, and mice were compared based on genotype and gender. Additionally, serum was collected from Cftr-/-(11 male and 11 female), Cftr+/-(9 male) and Cftr+/+ (13 male and 13 female) mice, also at 14 wks of age. Evidence of bone resorption was determined by measurement of C-terminal telopeptide α1 chain of type 1 collagen (CTx). Results: Total metaphyseal bone density was reduced in male Cftr

Female Cftr-/-mice were found to have reduced metaphyseal bone density compared to Cftr+/-mice. Trabecular density was reduced in both male Cftr+/-and Cftr-/-mice, as well as female Cftr+/-mice, compared to Cftr+/+ controls. Greater concentrations of CTx were discovered in Cftr-/-mice compared to Cftr+/+ (36.1±6.7 vs. 16.9±1.3 ng/mL, p<0.01), but there was no significant difference between Cftr+/-mice (22.1±2.2) compared to either Cftr-/-or Cftr+/+. Conclusions: Heterozygote Cftr+

However, there is no significant difference in type 1 collagen breakdown products for heterozygote mice, implying an etiology other than increased bone resorption for the partial bone phenotype. Additionally, these findings in CFTR-deficient and heterozygote mice occur in the absence of additional clinical confounders and suggest a direct effect by the Cftr genotype

Metaphyseal Total and Trabecular Bone Density *p<0.05 compared to Cftr+/+, ***p<0.001 compared to Cftr+/+ and p<0.05 compared to Cftr+/-568ଙ

5%) more than one. The radial bone was characterized by an inadequately thin cortex in relation to muscular force, as shown by a significantly reduced height-adjusted cortical thickness at the proximal radius (-1.09 ± 1.22 SDS) and by a reduced relative cortical area (-1.22 ± 1.60 SDS), while the mineralization of radial trabecular bone was unaltered. As a consequence of cortical thinning, both the age-adjusted (-1.56 ± 0.94) and the height-adjusted (-0.80 ± 0.82 SDS) Strength-Strain Index that reflects the combined strength of trabecular and cortical bone, was reduced. In addition, the height-adjusted muscle crosssectional area (-1.34 ± 0.89 SDS) was significantly decreased. BMD by DEXA both of the lumbar spine (-1.38 ± 1.21 SDS) and the total skeleton (-0.67 ± 1.14 SDS) was also compromised. The most prominent biochemical alteration was vitamin D insufficiency (25(OH)-D level <20 ng/ml) in 47%, leading to secondary HPT in 11%. Conclusions: The bone

Cystic fibrosis (CF)-related bone disease is associated with reduced bone mineral density (BMD), increased fracture risk, reduced bone formation and enhanced osteoclastic bone resorption. Chronic illness, malnutrition, vitamin D deficiency, glucocorticoids, inflammatory cytokines and hypogonadism contribute to CF bone disease. However, epidemiological investigations and knockout animal models suggest a direct link between CFTR inactivation in bone and low BMD. Studies were performed examining mechanisms of bone cell dysregulation due to CFTR inactivation. Osteoblasts expressed significantly more CFTR than osteoclasts, so studies were directed at this cell type. In the calvarial organ culture assay, Cftr null mice had reduced bone formation compared to WT littermates (6430 µm 2 vs. 16440 µm 2 , p=0.0023). Subconfluent Cftr null calvarial osteoblasts expressed substantially less differentiation markers (Osterix, Collagen 1a1 and Osteocalcin) than WT osteoblasts indi

PTH-mediated osteoblast PTH1R internalization was reduced (p<0.0001) suggesting that CFTR inactivation may impair normal bone remodeling by dysregulating osteoblast PTH1R recycling and/or signaling. We present evidence that uncoupled bone turnover of CF bone disease is a direct consequence of osteoblast CFTR inactivation by delaying osteoblast differentiation and enhancing osteoclastogenesis via reduced Opg expression. These data bolster the need for research comparing anabolic vs. antiresorptive treatments for CF bone disease. adolescents with CF

Brief psychosocial screening in outpatient pediatric practice

The Children's Depression Inventory (CDI)

Patterns of children's coping with an aversive dental treatment

Consulting Psychologists Press. health status, and investigating which types of exercise are most beneficial for adolescents with CF. Funding was provided by NIH grant (R01 HL #47064). References: 1

Performance Improvement, MD Anderson Cancer Center

Children's Harbor Family Center

Perkett, E. 1 1. Vanderbilt Children's Hospital

Methods: A subcommittee was established by the CF Foundation Center Committee to gather information about current involvement of NPPs. A survey was sent to adult, pediatric and affiliate CF program directors (PDs) using the CFF database. A second survey was sent to NPPs from CFF contact information and referrals from PDs. Results: PD survey: Responses were received from 108 PDs (49% pediatric, 34% adult, 17% affiliate) with center sizes ranging from 30-700 patients (mean 177 +/-128). Overall, 57 (53%) programs had NPPs and 76 (70%) had or planned to hire NPPs. Academic programs comprised 84 (77%) of programs. Of the 57 programs with fellows, 58% had NPPs. PDs described collaborative relationships between fellows and NPPs. Reasons for NPP use included ideal clinical role (75%), expansion of services (72%) and physician shortage (40%). Programs actively recruiting physicians Based on the survey results, education related to infection and infection control measures continues to be justified. Staff need to be educated about use of gowns and gloves with all CF patients to assure 100% compliance with our standard. Parent/child education regarding signs of possible lung infection needs to be offered frequently

A. 1,11 ; Braggion, C. 2

Clinical conditions of CF patients upon recruitment: Pancreatic insufficiency was present in 53 cases (86.9%) and CFRD in 24 (39.2%); mean FEV1 49.4% (sd 20.2); mean BMI 20.4 kg/m 2 (sd 3.2). TIVAD features: the thoracic area was chosen in 78.7% of cases for placing the port. Single reservoir TIVADs with silicone catheter were used most frequently. Veins used were subclavian (34.4%), jugular (39.3%), others (26.3%). Reasons for TIVAD positioning: reasons for TIVAD positioning were lack of superficial veins in 93.4% of cases, need to carry out home therapy in 31.1%, repeated access to hospital in 6.6%, infusion of hypertonic medications in 1.6% (not reciprocally exclusive). Clinical conditions of CF patients upon insertion: mean BMI 19.8 kg/m 2 (sd 2.9); mean FEV1 47.31% (sd 18.2)

Salary support for NPPs included hospital support (67%), billing (39%), center grant (35%), and other grant/contract (25%). NPPs bill for outpatient and inpatient care in 65% and 28% of programs, respectively. Many PDs did not know about NPP billing with 5% and 26% reporting no knowledge of outpatient and inpatient billing respectively

research (81%) and leadership (55%) were identified. Benefits of NPPs included access to care, approachability and continuity of care. Obstacles identified were billing; training; uncertainty of the NPP's role by families, CF team and NPPs themselves; and unclear or too broad scope of practice with many roles in addition to clinical care roles. Conclusions: NPPs are working with physicians in many centers and have the potential to help meet the increasing clinical workforce need. Further evaluation of financial issues is indicated to continue the support of NPP jobs in CF. Roles and expectations need to be clearly defined. Initial and ongoing training standards and opportunities should be explored

Israel Patients with cystic fibrosis (CF) require frequent IV antibiotic therapy to treat acute pulmonary exacerbations. Peripheral veins may be damaged after multiple insertion of IV placements and/or be difficult to locate. Therefore permanent catheters are introduced, often for the lifetime of the patient. These devices are widely used in the management of CF patients as they increase substantially their quality of life. Aim: To analyze the safety and efficacy of indwelling catheters in our Center

28 PICC-Lines, 14 Port-A-Cat and 4 Hickman lines. Central-line derived complications developed in 11 patients as follows: 10 occlusions in 4 patients (5 in the same patient due to non-flushing of the catheter), 6 infections in 3 patients and 3 thrombi in 3 patients detected by ultrasound. One of 3 subsequently developed pulmonary embolism diagnosed by CT, requiring hospitalization due to deterioration in his saturation

In 3 patients the PICC -Line was removed after 14 days, due to discomfort. Conclusion: Overall the use of permanent catheters was found to be safe

Children's Hospital of Pittsburgh of UPMC

Acknowledgements: This work was supported by a grant from the National Institutes of Health, DK 074010 to SMO, PI. Katherine Schiller was supported on a NHLBI Pulmonary & Critical Care training grant through the center for Lung Science and Health, University of Minnesota. Acknowledgements: Funded by CFF EPIC09K0. Acknowledgments: This study was supported by Genentech, Inc., South San Francisco, CA. The authors gratefully acknowledge the patients, parents, investigators, and coordinators of the ESCF.A number of previous investigations have reported that cystic fibrosis (CF) patients with liver disease have worse FEV1 than those without liver disease, whereas others have reported no difference or even a slower annual rate of FEV1 (% Pred) decline. We speculate that the lack of a welldefined relationship between liver disease and pulmonary function in CF reflects, in part, the lack of a uniform definition of clinically relevant liver disease. To assess the relationship between clinically significant liver disease and pulmonary function, we retrospectively evaluated spirometric data from 235 CF patients (62% male) with severe liver disease (CFLD) as defined by the presence of portal hypertension (confirmed by splenomegaly, esophageal varices, and/or hypersplenism). The distribution of FEV1 (% Pred) values in this cohort (mean age 11.2 yrs +/-5.4) at or before the diagnosis of CFLD approximated a normal distribution (confirmed by Shapiro-Wilk test, p=0.28), indicating no significant shift toward mild or severe lung disease. We also compared those FEV1 measures obtained at the time of (or prior to) the diagnosis of CFLD to CF-specific reference values matched for age and gender, as defined by Kulich percentile; there was a strong correlation between FEV1 (% Pred) and Kulich percentile (r=0.85; Figure 1 ). Lastly, rates of annual decline of FEV1 (% Pred) before diagnosis of CFLD showed no significant difference when compared with post-CFLD rates of decline (p=0.34). Taken together, these results suggest that pulmonary function at the onset of CFLD (defined by portal hypertension) is neither more mild nor severe than that of matched CF patients without CFLD, suggesting no strong overlap for the underlying genetic risk in these two diseases.

(mean±SD; 4.5±0.8) in adults, ages 18-24 and ≥25 yrs, was determined from ESCF data collected from 1994-2005. Risk factors for decline (including signs and symptoms, medical conditions, and microbiology), were identified and compared among and within age groups with repeated-measures, mixed model regression.Results: Mean (±SD) baseline FEV1 % pred was 67.7±22.0 for ages 18-24 (n=2795) and 55.7±21.0 for ages ≥25 yrs (n=1378). The mean rates of FEV1 decline (% pred/yr) were -1.93 and -1.35 respectively. Independent risk factors for decline in the 18-24 yr group (p < 0.05, ranked in order of impact on decline) were pancreatic enzyme use (surrogate for pancreatic insufficiency)(-0.88 % pred/yr), B. cepacia (-0.85 % pred/yr), multiply resistant P. aeruginosa (-0.61 % pred/yr), asthma (-0.32 % pred/yr), sputum (-0.31 % pred/yr), mucoid P. aeruginosa (-0.30 % pred/yr) , and female sex (-0.28 % pred/yr) . Three factors were identified as contributing to less decline, wheeze (0.55 % pred/yr), sinusitis (0.41 % pred/yr), and baseline FEV1 % pred <40 (1.19 % pred/yr higher than the 40-69 group). Restricting the analysis to those 18-24 yrs with FEV1 % pred ≥40 yielded similar results (risk factors and impact), and identified an additional risk factor (cough; -0.34 % pred/yr), while baseline FEV1 % pred, sputum, wheeze and asthma were no longer significant. For the ≥25 yr group, the only significant risk factor for decline was pancreatic enzyme use (-0.53 % pred/yr), while baseline FEV1 % pred <40 contributed to less decline (0.72 % pred). Restricting the analysis to those ≥25 yrs with FEV1 % pred ≥40, Aspergillus was identified as a risk factor (-0.67 % pred) , the impact of pancreatic enzyme was stronger (-0.57 % pred/yr), and sinusitis was identified as contributing to less decline (0.52 % pred/yr). The ≥25 yr group was divided and analyzed for ages 25-31 (n=772) and ≥32 (n=606). Results were similar between these 2 age groups. Compared to our previous analysis in children (ages 6-17 yrs) with CF (Konstan et al, J Pediatr 2007; 151:134-9) , our analysis in adults differed for several risk factors. Most notable was no effect of baseline FEV1 % pred ≥40, and the exclusion of crackles and IV-treated exacerbations as risk factors in adults.Conclusions: This study identifies and assesses the impact of risk factors for FEV1 decline in adults with CF. Many of the risk factors identified for the youngest adults (ages 18-24 yrs) were similar to those identified in our previous work in children. We speculate that a survival effect in older adults (≥ 25 yrs) accounts for the differences in risk factors. As with our previous work in children, the results from this study in adults might allow clinicians to determine the estimated risk of FEV1 decline for an adult with CF. Acknowledgement: Supported by Genentech, Inc.

VanDevanter, D. 1 ; Wagener, J. 2 ; Pasta, D. 3 ; Elkin, E. 3 ; Jacobs, J. 4 ; Morgan, W. 5 Rationale: Loss of lung function in cystic fibrosis (CF) patients is associated with increased mortality and varies among individuals and over time. Younger children can experience disease progression that is not easily identified and traditional spirometry, the standard by which CF lung disease progression is measured, cannot be reliably performed under age 6. A previous analysis using data from the Epidemiologic Study of CF (ESCF) showed relationships between clinical variables at age 3 and lung function at age 6 ( Konstan et al., J Pediatr 2003; 142:624-30) . We have extended this work using ESCF data to develop a simple tool for use during a routine clinic visit for children ages 2-5 yrs to estimate lung function at age 6.Methods: Subjects ages 2-5 yrs were included if they had an index clinic visit (ICV) prior to 1/1/1998 and >1 yr after ESCF enrollment. FEV 1 % predicted measured at a stable follow-up clinic visit between the 6 th and 7 th birthday was used as the outcome. Age, sex, sputum production, cough, wheeze, clubbing, crackles, sinusitis, elevated liver enzymes, pancreatic enzyme use, weight-for-age percentile, height-for-age percentile, number of past year exacerbations, and culture status at the ICV were used to predict FEV 1 % predicted in multivariable linear regressions. Variable interactions were also studied. A "risk score" was created by assigning whole numbers to risk factors with each 1-unit difference representing a 2% predicted deviation from 100% predicted at age 6. The relationship of the risk score and actual FEV 1 % predicted at age 6 was assessed by Pearson correlation coefficient.Results: 2,709 CF patients were included in the analysis. Whole numbers were assigned to categories for cough, wheeze, clubbing, crackles, weight-for-age percentile, number of past year exacerbations, and P. aeruginosa infection status and summed to create a risk score ranging from -18 to +2 (theoretical range, -20 to +2), with a median score of -1 (mean, -2.1; standard deviation, 3.3) . No variable interactions remained in the final algorithm. The relationship between risk scores and FEV 1 at age 6 approximated target values, with a risk score of 0 equal to 99.3% predicted FEV 1 and each score unit equal to 2.09% change in predicted FEV 1 . The Pearson correlation coefficient was +0.35. For any given risk score, mean and median FEV 1 % predicted for patients with that score increased as the value of the score increased.Conclusions: This simple category scoring tool to predict FEV 1 % predicted at age 6 uses information available at any clinic visit. The tool has the potential to identify young patients at high risk of having poorer lung function at age 6. The results might better inform clinicians and families about the risk of measurable lung disease by age 6 yrs in young children with CF.Background: The Cochrane Cystic Fibrosis & Genetic Disorders (CFGD) Review Group was established in 1995 and publishes systematic reviews of interventions for cystic fibrosis (CF), which are available via The Cochrane Library (www.thecochranelibrary.com). The Editorial Board includes 5 CF specialists. A Cochrane review provides systematic evaluation of the best evidence available to determine the effectiveness of an intervention (for example, hypertonic saline) for people with CF. A review will generally have 3-5 authors; international collaborations are encouraged. Following publication of a peer-reviewed protocol, authors complete the review (often with support and guidance from the Editorial base staff) and after successful open peer review (including statistician and consumer) the review is published.Aim: To assess the impact of Cochrane CF reviews on the care of people with CF.Methods: The most recent edition of the Cochrane Library (2009) was examined for CF reviews and protocols. We noted the number of trials from the CFGD register (a comprehensive database of CF-related clinical trials) that have been evaluated. We examined citations in national guidelines and feedback from authors.Results: Forty-five reviews and 6 protocols directly related to CF have currently been published in the Cochrane Library by the CFGD. In December 2008, a total of 1072 CF trials were listed on the CFGD register; 478 (45%) of these have been evaluated critically and 178 (16%) were of sufficient quality to be included in published systematic reviews. All reviews contain recommendations for future research and in five cases this had a direct impact resulting in new trials. Data from these trials have then been incorporated in the updated reviews leading to a change in recommendations for clinical practice. Reviews have also been cited in consensus documents and guidelines published by the UK CF Trust, the Cystic Fibrosis Foundation, the European Cystic Fibrosis Society and the Thoracic Society of Australia and New Zealand. The Cochrane Database of Systematic Reviews now has an impact factor of 4.654 and is ranked 14th out of 100 journals in the 'Medicine, General and Internal' category.Conclusions: Systematic reviews have a direct impact on practice and can also identify gaps in current evidence and help set the research agenda. Reviews are written by volunteers, usually healthcare professionals, with support from the Editorial Base. Anyone who is interested can contribute as an author, peer reviewer or consumer referee.

Wagener, J. 1 ; McColley, S. 2 ; Elkin, E. 3 ; Yegin, A. 4 ; Pasta, D. 3 ; Liou, T. 5 ; Morgan, W. 6 ; Konstan, M. 7 1. University of Colorado Denver School of Medicine, Aurora, CO, USA; 2. Northwestern University Feinberg School of Medicine, Chicago, IL, USA; 3. Icon Clinical Research, San Francisco, CA, USA; 4. Genentech, Inc., South San Francisco, CA, USA; 5. University of Utah, Salt Lake City, UT, USA; 6. University of Arizona, Tucson, AZ, USA; 7. Rainbow Babies and Children's Hospital and Case Western Reserve University, Cleveland, OH, USA Rationale: Determining quality outcomes in patients with cystic fibrosis (CF) has historically relied on ranking care sites based on patient data from a single year (Quinton and O'Connor. Clin Chest Med 2007; 28:459-72) . This site ranking also correlates with practice patterns (Johnson et al. Chest 2003; 123:20-7) . However, change in pulmonary function over time (slope) may be a more important metric to determine best practice. Therefore we used data from the Epidemiologic Study of Cystic Fibrosis (ESCF) to determine how site rankings differ based on various single-year measures vs. multi-year measures including FEV1 slope.Methods: Data from ESCF for the years 2000-2002 were used to identify CF care sites with at least 10 patients in one of three age groups (6-12, 13-18, ≥18 years). Patients had to have received care at that site for all 3 years and have results from at least two "clinically stable" pulmonary function tests (PFTs) per year. Sites were ranked by the median of their patients' PFT values (separately for FEV1, FVC, or FEF25-75) using three methods: the best value in 2002 (single-year); the average of each best value in 2000, 2001, and 2002 (3-year average) ; and the slope of best values from 2000 to 2002 estimated by simple linear regression (3-year slope) . Spearman rank correlation coefficients were used to compare the site rankings by the different methods.Results: A total of 113 sites were included. The site median FEV1 in 2002 correlated well with other PFT variables for that year as well as the 3-year average FEV1 of the best value from each year (Table) . There was a much smaller correlation with the slope over the prior 3 years. Because patients with severe disease often demonstrate little decline in FEV1 we re-analyzed site ranking based upon only those with FEV1 ≥50% predicted. This minimally altered the results with a minor increase in the correlation of single-year FEV1 with 3-year slope in the ≥18-year-old group (r = 0.19).Conclusions: These results suggest that using single-year, or 3-year average, patients' PFTs to rank sites does not represent the rank based on the PFT slope over 3 years. We speculate that using PFT slope over several years may better reflect the impact of care within a site and should be considered when evaluating best practices.

The authors gratefully acknowledge the patients, parents, investigators, and coordinators of ESCF. Supported by Genentech, Inc.

Sufian, B.S.; Passamano, J.A. Sufian & Passamano, L.L.P., Houston, TX, USA Introduction: Maintaining health insurance is an important concern to CF patients. In the past year, most likely due to the economic recession, the number of patients with problems related to health insurance greatly increased. The issues are: (1) losing employer-sponsored health insurance; (2) needing to extend coverage under COBRA; (3) seeking to qualify for Social Security Disability Insurance; (4) retaining coverage under a parent's plan as a disabled child; (5) obtaining Supplemental Security Income; and (6) losing benefits due to marriage.Method: The CF Legal information Hotline is sponsored by the CF Foundation. It began operation in 1998 and provides free and confidential legal information to people with CF. It has received over 12,000 calls in 11 years from throughout the U.S. The calls received by the Hotline reflect the legal issues affecting the CF community. The average number of calls per month by subject in 2009 is compared to 2008 to determine any significant change in problems experienced.Results: The study showed a significant increase in concerns about maintaining health insurance. It showed a 117.8% increase in the number of persons with CF or their parents who lost employer-sponsored health insurance. The number of callers seeking to extend coverage under COBRA increased 153.2%.The study showed a 70.2% increase in the number of persons with CF seeking Social Security Disability Insurance. Many of these are young adults who experienced a decline in health after entering the workforce. Yet, no significant increase was observed in the numbers seeking to qualify for Supplemental Security Income (SSI).The study showed a 35.3% increase in the number of persons with CF who reached a limiting age on their parents health plan, but sought to remain on the plan because they are incapable of self-support. These individuals are young adults completing high school or college. Finally the study shows an 87.1% increase in the number of young adults concerned about how a proposed marriage will affect their health benefits.Discussion: Persons who lost their jobs were often eligible for a COBRA subsidy which made COBRA affordable. Direction as to coverage options was helpful. Young adults reaching a limiting age under their parents plan are aided by Disabled Child laws that allow a child incapable of selfsupport to remain a dependent on a parent's plan.Workers unable to work seeking SSDI benefits have several concerns, including maintaining coverage during the 29 month Medicare waiting period and the 5 month income waiting period. Others experience difficulty when medical records do not adequately reflect the conditions of eligibility. Some report tension between a patient's need to stop work and the care team's use of employment as a measure of successful treatment.Marriage may adversely affect health benefits. Marriage can cause a person to lose coverage under a parent's policy, without adequate replacement coverage. The combined assets of spouses may cause the spouse with CF to become ineligible for Medicaid or cause the loss of Medicare subsidies.Conclusion: Obtaining and maintaining insurance coverage remains the single most important legal concern among CF patients and families. Coverage issues increase and are harder to solve during a recession. Background: Acute treatments such as intravenous (IV) antibiotics are used mainly for treatment of bacterial infections that result in cystic fibrosis (CF) exacerbation admissions. In the 2008 CF Exacerbation guidelines, the CF Foundation recommends the use of once-daily IV aminoglycosides in all CF patients with acute pulmonary disease. In early 2006, we started using the once-daily tobramycin dosing strategy for most CF patients. We monitored toxicity and efficacy via a trough drawn 30 minutes prior to the second dose and a peak drawn one hour after the second dose: Trough Dose Peak (TDP). Dosing adjustments were made using standard pharmacokinetic equations or by adjusting the dose by 10% in order to achieve a peak of 20-30 mg/mL and a trough < 1 mg/mL. This process was performed until 10/2008 when a change in the monitoring strategy was initiated based on information from the 2008 NACFC. We began to use a peak drawn one hour after the second dose and a random level drawn at eight to twelve hours: Peak Random (PR). In addition, target ranges of Cmax (20-30 mg/mL) and AUC (80-125 mg/L/hr) were used to guide therapy. The objective of this study is to evaluate the initial effects of the change in tobramycin pharmacokinetic monitoring strategy.

Methods: We conducted a retrospective data analysis of all CF admits utilizing once-daily tobramycin therapy from 01/01/2007 to 03/31/2009. Descriptive statistics and student's t-test were used for data analysis. The variables included were dose in mg/kg (before/after the change in monitoring process), half life (T1/2) (before/after), volume of distribution (Vd) (before/after), Cmax (after), AUC (after), number of tobramycin laboratory draws per admit (before/after), number of dose changes per admit (before/after), and serum creatinine values (before/after). This project was IRB-approved.Results: A total of 79 patients with 129 admissions were included in the study: 54 patients with 100 admissions (TDP) vs. 25 patients with 29 admissions (PR). The mean mg/kg/dose was 10.2 mg/kg/dose (TDP) vs. 10.3 mg/kg/dose (p=NS) (PR). The mean T1/2 was 3.3 hours (TDP) vs. 1.89 hours (PR) (p<0.001). The mean Vd was 13.85 L (0.35 L/kg) (TDP) vs. 9.9 L (0.3 L/kg) (PR) (p=0.001). The mean calculated Cmax for the PR group was 32.8 mg/mL. The mean calculated AUC for the PR group was 96.5 mg/L/hr. There was a mean of 3.6 tobramycin levels per admit with 1.5 dose changes (TDP) vs. 4 tobramycin levels per admit with 1.6 dose changes (PR) (p=NS). The mean serum creatinine values drawn during admit were 0.62 mg/dL and 0.58 mg/dL, respectively (TDP) vs. 0.52 mg/dL and 0.5 mg/dL, respectively (PR).Conclusion: A statistically significant difference was found between the mean T1/2 and Vd between the TDP and PR methods, however, this did not translate to a change in overall dosing. The PR method calculated a T1/2 within the published aminoglycoside half-life range in CF patients (1-2 hrs) . The TDP method was not as reliable at predicting the T1/2 and Vd possibly due to the inability of the tobramycin assay to detect levels below 0.6 mg/L or the potential inaccuracy of using an extended interval trough to calculate elimination. Healthcare systems have not universally adopted systematic approaches to quality improvement (QI). Using Cystic Fibrosis Foundation Learning and Leadership Collaborative tools for enhancing delivery of care, the Central CT CF Center's (CCCFC) global aim is to improve adherence and selfmanagement of care in patients with cystic fibrosis (CF). We hypothesize if adherence and co-management improve, disease management improves, and potentially so will health outcomes.

Background: Pancreatic involvement is common and the injury is progressive in cystic fibrosis (CF), leaving over 85% of patients with pancreatic insufficiency (PI). Despite a universal involvement of the pancreas in CF, the mechanisms leading to its destruction are not well-known. CFTR gene knock-out mouse models do not recapitulate the pancreatic involvement characteristic of CF in humans.Hypothesis: Pancreatic involvement in a CFTR-/-pig model will be similar to humans with CF.Methods: We investigated the pancreatic histology and the expression of exocrine and endocrine cell markers in CFTR-/-newborn pigs and contrasted our findings with CFTR+/+ pigs. We used microarray expression profiling to explore the differences in gene transcription between CF and non-CF pig pancreata using an Affymetrix Porcine GeneChip.Results: Lobular exocrine tissue was significantly reduced in CFTR-/pancreata with variable degrees of acute and chronic inflammation. There were scattered neutrophils and macrophages infiltrating the acini and the inspissated material in ectatic ducts. In contrast, lymphocytes and plasma cells infiltrated the interstitial space; they formed large aggregates in areas of severe exocrine tissue destruction often centered on the ectatic remnant ducts. Pancreatic ducts were dilated with mucus surrounding centrally oriented inspissated zymogen secretions. Pancreatic amylase, lipase and trypsin activities, and pancreatic elastase and chymotrypsinogen expression were markedly decreased in CFTR-/-pigs. Immunohistochemistry staining for insulin, glucagon and somatostatin were present and comparable in all pigs. Microarray expression profiling showed unchanged mRNA expression for glucagon and somatostatin, diminished mRNA expression for pancreatic exocrine enzymes, and upregulation of proinflammatory genes in CFTR-/-pig tissue. Interestingly, mucin producing gene expression (MUC1, MUC5B, MUC5AC) was diminished in CFTR-/-pig pancreata compared to CFTR+/+ pigs.Conclusion: CFTR-/-pig pancreas had increased inflammatory cell infiltration and proinflammatory gene expression compared to CFTR+/+ pigs. It is currently unknown whether inflammation precedes the pancreatic lesions or develops in response to tissue destruction. Our studies show that the pancreas is severely involved in CFTR-/-pigs, similar to humans with severe CFTR mutations. CFTR-/-pigs may be an excellent model to study the mechanisms underlying pancreatic disease in CF. 5 1. Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA; 2. Indiana University School of Medicine, Indianapolis, IN, USA; 3. Penn State Milton S. Hershey Medical Center, Hershey, PA, USA; 4. Solvay Pharmaceuticals, Inc., Marietta, GA, USA; 5. Solvay Pharmaceuticals GmbH, Hannover, Germany Introduction: Exocrine pancreatic insufficiency (EPI) occurs in 85-90% of subjects with cystic fibrosis (CF). For these patients, pancreatic enzyme replacement therapy (PERT) is critical to achieve adequate nutrition, and is considered standard of care.

Objective: To compare a new formulation of pancrelipase delayedrelease 24,000-lipase unit capsules (study drug) in subjects with EPI due to CF according to age and severity of EPI.Methods: Prospectively planned exploratory subanalyses from a double-blind, randomized, multicenter, placebo-controlled, two-period crossover study that demonstrated benefits of study drug in 32 subjects with CF ≥12 years of age with confirmed EPI. Patients were randomized to one of two sequences, pancrelipase then placebo, or placebo then pancrelipase (5 days each with intervening washout period of 3-14 days on their usual PERT). We measured the coefficient of fat absorption (CFA), the coefficient of nitrogen absorption (CNA), clinical symptomatology, and safety parameters by age (12-18 years vs >18 years of age) and severity of EPI (placebo CFA ≤ 50% [more severe] vs CFA >50% [less severe]).Results: The effects of study drug on CFA and CNA (placebo-adjusted change from baseline) were slightly higher in subjects 12-18 years of age compared to those >18 years of age. LS means ± standard error (SE) for treatment difference from placebo were CFA 43.4 ± 5.7% vs 37.3 ± 4.2%, respectively and CNA 39.0 ± 5.2% vs 33.6 ± 3.3%, respectively (all P<0.001 vs placebo). Subjects with more severe EPI showed greater improvement with study drug compared to those with less severe EPI: CFA and CNA LS means ± SE for treatment difference were 52.4 ± 2.5% vs 23.3 ± 2.9%, respectively and 42.3 ± 3.0% and 26.6 ± 3.9%, respectively (all P<0.001 vs placebo). However, the active-treatment LS means ± SE at the end of the study drug period were similar (CFA 88.7 ± 1.8% vs 88.7 ± 2.0%, respectively; CNA 83.7 ± 2.1% and 86.3 ± 2.7%, respectively). Similar patterns were observed for symptom improvement on study drug vs placebo by age (comparable in both age groups) and by EPI severity (greater improvement in subjects with severe EPI) although sample sizes were too small to reach meaningful conclusions. Fewer treatment-emergent adverse events (TEAEs) were reported on study drug than placebo irrespective of age and EPI severity. Study discontinuation due to the TEAE weight loss occurred in 1 patient one day after last day of study drug in the first cross-over period. There were no serious TEAEs and no differences between treatments in laboratory abnormalities.Conclusion: In this study, pancrelipase delayed-release 24,000-lipase unit capsules were effective in treating EPI due to CF and had a tolerable safety profile. Both younger and older patient groups experienced similar changes in CFA and CNA. Those with more severe EPI demonstrated greater improvement.Supported by Solvay Pharmaceuticals Inc., Marietta, GA, USA. Introduction: Children with exocrine pancreatic insufficiency (EPI) due to cystic fibrosis (CF) require pancreatic enzyme replacement therapy (PERT) in order to maintain adequate nutrition. Pancrelipase delayedrelease capsules have provided this essential therapy; however, in response to recent regulatory mandates, all PERT products seeking US FDA approval require clinical efficacy and safety studies.

Objectives: Our aim was to demonstrate superior efficacy of a new formulation of pancrelipase delayed-release 12,000 lipase-unit capsules vs placebo in children with EPI due to CF.Methods: This was a double-blind, randomized, placebo-controlled, multi-center, two-period cross-over study in children aged 7-11 years with a confirmed diagnosis of CF and EPI. Patients were randomized to one of two treatment sequences for a period of 5 days: pancrelipase then placebo, or placebo then pancrelipase, with a washout interval of 3-14 days in between on their usual PERT. Dosing was targeted at 4,000 lipase units/g of dietary fat intake. Each patient received a prospectively designed diet to include at least 40% of calories derived from fat. The primary outcome measure was the coefficient of fat absorption (CFA); the study also investigated the coefficient of nitrogen absorption (CNA), stool output, clinical symptoms, and safety parameters.Results: Seventeen subjects were randomized (8 placebo then pancrelipase, 9 pancrelipase then placebo). One subject (pancrelipase then placebo group) withdrew consent in the first treatment period; 16 subjects completed the study. The median age was 8.0 years, 12 (70.6%) were male, and the mean daily lipase dose was 4,472 units/g fat consumed. CFA values were significantly greater on pancrelipase vs placebo, with least square (LS) means ± SE of 82.8 ± 2.7% vs 47.4 ± 2.7%, respectively (p<0.001). Similar results were observed for the CNA: 80.3 ± 3.2% vs 45.0 ± 3.2% (p<0.001). Stool output was significantly reduced with pancrelipase vs placebo: LS means ± SE for 72-hour stool weight (g) 166 ± 28 vs 436 ± 33 (p<0.001); 72-hour stool fat (g) 58 ± 10 vs 183 ± 10 (p<0.001); and stool frequency per day 1.9 ± 0.2 vs 3.4 ± 0.2 (p<0.001). Symptoms of EPI (abdominal pain, flatulence, and stool consistency) were improved with pancrelipase compared with placebo. Treatment-emergent adverse events (TEAEs) were reported in 5 (29.4%) patients taking pancrelipase vs 9 (56.3%) receiving placebo and were predominantly gastrointestinal events. There were no treatment discontinuations due to TEAEs, or serious TEAEs reported, in any patients receiving pancrelipase or placebo.Conclusion: In CF children with EPI, this new formulation of pancrelipase dramatically improved CFA, CNA, stool output and symptoms, and was safe and well tolerated in this study.Supported by Solvay Pharmaceuticals, Inc., Marietta, GA, USA.

There was a significant association between severity of CFTR mutations and development of pancreatitis, with patients carrying mild mutations more likely to develop pancreatitis (Table) . The odds ratio for developing pancreatitis with mild mutations was 2.2 (95% CI, 1.2-3.9; P=0.006) compared with intermediate and severe mutations. Patients with CFTR-PANC developed pancreatitis at a younger age (median, 15.7; range, 2.1-36.0) compared with 22.8; range, P=0.009) . There was no significant difference in gender ratio between the 3 groups. Overall, there was a difference in sweat chloride and FEV1 % predicted values between the 3 cohorts but these differences were not significant between CFPS-PANC and CFPS-NOPANC groups.Conclusion: There is an increased risk of developing pancreatitis in patients who carry functionally mild mutations and those who are in the mildest end of the spectrum of CFTR dysfunction develop pancreatitis at a younger age. Future studies are necessary to understand these observations. Apart from the risk of pancreatitis, there was no phenotypic difference between patients with CFPS-PANC and CFPS-NOPANC. Background: Australian guidelines for administration of pancreatic enzyme replacement therapy (PERT) recommend that enzymes should be distributed over the day based on the fat content of meals and should be taken before, or before and during the meal (1). These recommendations are however, consensus rather than evidenced based. Limited data exists as to how the timing of PERT in relation to a meal affects efficacy. Taylor et al. (2) showed that there was a mismatch between gastric emptying of PERT and a meal, with a mean difference of 67 minutes (range 3-129 minutes). Hence there may be a benefit in taking enzymes after a meal. The aim of this pilot study was to investigate whether PERT given before or after a meal resulted in improved lipase activity.

Methods: Ten children with CF aged 10.2 ± 1.25 years (mean ± SEM; range 5-16 years) completed two 13 C-Mixed Triglyceride (MTG) breath tests after administration of PERT either 10 minutes before or after a test meal to assess lipase activity and one 13 C-Octanoate breath test (OBT) to assess gastric emptying rate. The labeled isotope was ingested combined with a test meal after an overnight fast. Breath samples were collected and analysed for 13 CO 2 content. Results: On an individual level we were able to show significant improvements in the cumulative dose recovered (PCDR) of 13 C-MTG when the pattern of PERT was switched to either before or after the meal. Five subjects had a greater PCDR when PERT was taken 10 minutes before the test meal and five subjects when PERT was taken 10 minutes after the test meal. Improvements in the PCDR ranged from 0.69 -21.58%.The improvement in PCDR as measured by highest PCDR minus lowest PCDR was not correlated with gastric emptying rate (r= -0.31, p= 0.39).Conclusion: This small pilot study indicates that on an individual level there is a wide variation in PERT efficacy and that timing has an impact with 6 out of 10 patients having an improvement in PCDR of greater than 10% when the PERT pattern was changed. Clinically, improvements in lipase levels by 10% or more could result in improved nutritional outcomes or the requirement for less PERT. Further research is required to determine why patients benefited from either of the two patterns. A larger scale study may be able to find a correlation with gastric emptying rates. The impact of nutritional status on quality of life (QOL) is unknown for children with cystic fibrosis (CF). In the present study, potential associations between early growth response, nutritional status and multiple domains of quality of life are examined over 3 years in children aged 9-19 years.Methods: Patients (N=100; mean age = 13.6 ± 2.8 years) were participants in the Wisconsin Randomized Clinical Trial of Cystic Fibrosis Newborn Screening. Beginning in 2002, the Cystic Fibrosis Questionnaire (CFQ) was administered yearly for approximately 3 years to assess QOL (observations=279). Seven CFQ dimensions were measured in both children and adolescents: physical functioning, emotional functioning, social, body image, eating disturbances, treatment burden and respiratory symptoms. These factors were standardized to a 0-100 point scale, with higher values representing better QOL. Height z-score (HtZ) and body mass index z-score (BMIZ), measured at the time of CFQ administration, and early growth response, characterized as the achievement of adequate weight gain within 2 years post-diagnosis (i.e., responder vs. non-responder), were evaluated using generalized estimating equations and included as covariates age, presence of meconium ileus at birth, pancreatic functional status and version of CFQ questionnaire.Results: Mean values of at least 80.0 were observed for physical functioning, emotional functioning, social, body image and eating disturbances. These dimensions also had the greatest proportion of scores equal to 100 (range 18% for social -62% for eating disturbances), and thus are subject to ceiling effects. Treatment burden (9% with score of 100, mean=65.2 ± 19.8) and respiratory symptoms (8% with score of 100, mean=77.5 ± 16.8) had the lowest mean values. No differences in CFQ dimensions were observed across the 3 yearly CFQ measures. HtZ was positively associated with eating disturbances (p=0.01) and BMIZ (p=0.01) was associated with higher scores for respiratory symptoms. Both HtZ and BMIZ, independent of each other, were positively associated (p<0.01) with physical functioning and body image. Early life growth response was associated with better scores for body image (p=0.01) and eating disturbances (p=0.045).Conclusions: Early growth response and concurrent nutritional status, as indicated by height and BMI z-scores, were positively associated with multiple dimensions of QOL in a population with high scores on most CFQ dimensions.(Supported by NIH Grants R01DK072126, R01DK34108 and Cystic Fibrosis Foundation Therapeutics grant.)Conclusions: PANCREASE-MT compared to placebo significantly improved fat and protein absorption that was accompanied by marked improvements in EPI symptoms, stool consistency and CGI scores. No unexpected adverse events were reported in this population of patients with CF. Background: Carnitine deficiency has been previously described in children with cystic fibrosis (CF) based on plasma levels; however, plasma carnitine levels neither convey tissue status nor functional information. Carnitine plays a central role in beta oxidation, which has not been assessed in CF using current technology.

Objective: To assess beta oxidation as a functional assay of carnitine adequacy and of mitochondrial function in children with CF and pancreatic insufficiency (PI).Methods: Baseline measurements from a multicenter nutrition intervention study comprising subjects with CF, PI and mild lung disease (FEV1 > 40% predicted) 6-18 years of age, and with no known liver or metabolic disease participated. Healthy subjects 6-18 years old served as a comparison group.Fasting plasma samples for both groups were analyzed by tandem mass spectrometry for acetyl (C2) and palmitoyl (C16) carnitine esters, and the C2:C16 ratio was calculated, as measures of carnitine adequacy. Statistical analyses were performed using Spearman rank correlations for associations among variables and analysis of co-variance (ANCOVA) for group differences adjusted for age and gender.Results: Seventy-four subjects (43 males, 10.9 ± 3.0 years old) with CF participated, and 190 children (95 males, 10.8 ± 3.3) served as a comparison group. The Table presents the medians [ranges] for the two groups for C2 and C16 carnitine esters, and for the C2:C16 ratio. In subjects with CF, the C2:C16 ratio declined significantly with age (r= -0.26, p=0.024). No significant age associations were found for the comparison group. After adjusting for age, gender and gender*group interaction, subjects with CF had significantly lower concentrations of C2 (p=0.038).Conclusions: Age affected acylcarnitine profiles in subjects with CF but not control subjects. These data suggest impaired flux through the fatty acid pathway and decreased acetyl CoA production which may be progressive with age, and may indicate an impairment of beta oxidation in children with CF and PI.Supported by NIDDK (R44DK060302), Clinical Translational Research Center (UL1RR024134), the Nutrition Center at the Children's Hospital of Philadelphia, and Avanti Polar Lipids, Inc. Low weight for age is an independent risk factor for decline in pulmonary function. Growth retardation of cystic fibrosis (CF) patients has been attributed to malnutrition from loss of exocrine pancreatic activity and cachexia from bacterial colonization of the lungs. Significant advancements had suboptimal levels of vitamin K; in another study, 90% of CF patients receiving vitamin supplementation remained vitamin D deficient. Two studies (n=151) demonstrated that CF subjects receiving supplemental vitamin A had elevated serum retinol levels, and this suggests that other sources of vitamin A such as beta-carotene be considered. Based on the data from the literature review, SourceCF multivitamins were reformulated with increased amounts of vitamin D and vitamin K and a greater ratio of vitamin A in the form of beta-carotene while keeping the total retinol activity equivalents the same as the original formulations.Conclusions: Vitamin needs of CF patients are unique; knowledge surrounding these needs is limited yet dynamic. As new evidence becomes available, CF-specific vitamin products require modification. SourceCF vitamin products were reformulated based on current nutrition evidence. Vitamins D and K and beta-carotene were increased, and retinol decreased. The SourceCF multivitamins also contain B-complex vitamins, biotin, ascorbic acid, and zinc. The new formulation should either reduce the need to use or reduce the doses of single vitamin supplements, such as D and K, to meet the vitamin needs of CF patients. The primary objective of the study was to evaluate the efficacy and tolerability of PANCREASE-MT capsules (MT 10.5 or MT 21) on the quantitative change in fat absorption in adults and adolescents/children with cystic fibrosis (CF) who had clinical symptoms of exocrine pancreatic insufficiency (EPI). Secondary endpoints included the coefficient of nitrogen absorption (CNA, a gauge for protein absorption), clinical signs and symptoms of EPI, stool consistency, and Clinical Global Impression (CGI) scores.

Methods: Following screening, subjects entered an open-label, run-in phase during which a high fat diet (approximately 100g/day) was initiated and current pancreatic enzyme therapy was replaced by PANCREASE-MT 10.5 or MT 21. When an optimal PANCREASE-MT dose was reached and the high fat diet was maintained for at least 3 days, subjects underwent the first 72 hour stool collection period to determine the open-label treatment coefficient of fat absorption (CFA). Subjects with a CFA ≥80% were randomized into the double-blind withdrawal phase to either continue on PAN-CREASE-MT or matching placebo. After a minimum of 24 hours in the double-blind phase, a repeat 72-hour stool collection was performed to determine the CFA of the double-blind phase. The CFA was determined by measuring both the fecal fat excretion and fat intake (g/day; a component of dietary data collection) during the stool collection periods.Results: Forty-nine subjects entered the open-label, run-in phase and 40 (14 adolescents 12 to <18 yrs; 26 adults ≥18 to 60 yrs) were randomized into the double-blind withdrawal phase. The mean CFA from open-label phase to the end of double-blind phase decreased slightly by 1.4% (88.2% to 86.8%) in subjects who continued to receive PANCREASE-MT compared with a decrease by 34.1% (90.5% to 56.4%) in subjects receiving placebo (between-group difference p<0.001). The mean CNA increased by 1.2% in the PANCREASE-MT group compared to a decrease of 26.6% in the placebo group (between-group difference p<0.001). Marked improvements were also noted in the clinical signs and symptoms of EPI (20% vs. 55%), stool consistency (10% vs. 70%, experienced greasy stools) and CGI scores: CGI-S-normal (85% vs. 55%); CGI-C-no change (80% vs. 35%); GAC-better (45% vs. 10%) in the PANCREASE-MT vs. placebo group, respectively. All treatment emergent adverse events were mild to moderate in severity and occurred less frequently in the PANCREASE-MT group (40% vs. 60%). The most common adverse events were gastrointestinal disorders including abdominal pain, diarrhea, flatulence, and abnormal feces.

Helm, J.M.; Webb, A.K.; Rowe, R.; Brennan, A.; Jones, A.M.

Background: Oral glucose tolerance testing (OGTT) followed by home serial capillary blood glucose (CBG) monitoring is the gold standard for determining glycaemic control in clinically stable adults with cystic fibrosis (CF) in the UK. However the OGTT only provides a "snapshot" of glucose handling at the time, classified as normal, impaired or diabetic glucose tolerance (NGT, IGT or DGT) according to World Health Organization (WHO) criteria. If abnormal, patients monitor serial CBG levels. CBG testing should be accurate to within 10% of corresponding plasma glucose levels, according to the American Diabetes Association (ADA). Continual glucose monitoring studies (CGMS) offer a novel approach in evaluating "real life" glycaemic control. A subcutaneous electrochemical sensor calculates interstitial glucose values every 5 minutes, producing a trace showing glucose trends for up to 72 hours. CGMS device calibration requires 4 CBG daily.Objective: We evaluated the accuracy of the CBG values compared with plasma glucose values, and CGMS glucose values compared with plasma glucose values, at the 0 and 2 hour time points of the OGTT.Methods: Twenty-seven clinically stable adults with CF commenced a 72 hour CGMS and attended the clinic for a 2 hour OGTT 1, 2 or 3 days later. They recorded 4 CBG daily during the monitoring period. During the OGTT, CBG and plasma glucose were taken immediately before and 2 hours after the standard glucose load (1.75g/kg up to maximum 75g). CGMS monitors were returned along with CBG results and downloaded sensor data was converted into a glucose trace using Solutions™ (Medtronic) software. The absolute and percentage deviation of each CBG value and CGMS value from its concurrent plasma glucose value was determined and summary statistics were performed.Results: Three of 27 (11.1%) CGMS sensors failed before OGTT, leaving 24 patients' traces (48 paired data sets) available for analysis. Two of 48 BM readings and 6/48 CGMS sensor values were technically incorrect and excluded from analysis. Twenty-nine of 46 (63.0%) CBG levels were within 10% of their paired plasma glucose values. The mean (SD) and median [range] deviation of CBG from paired plasma glucose values were 9.6 (7. There were 20 complete sets of patient data with valid plasma and CGMS values for both time points. By OGTT 8 individuals had NGT, 4 had IGT and 8 had DGT. The CGMS 2 hour time point value agreed with the conventional OGTT value in 15/20 subjects, underestimated glycaemia in 3 and overestimated glycaemia in 2 cases.Conclusion: Accuracy of CBG monitoring did not meet ADA standards but was "good" in keeping with reported data from clinical settings. CGMS single time point values were accurate within limits reported in the literature. These single time point CGMS values should not be interpreted alone but in conjunction with other available objective and subjective measures of glycaemic control. Further evaluation of CGMS traces is required to establish the most valid and reliable means of interpretation. Objective: In the setting of cystic fibrosis, both diabetes (CFRD) and impaired glucose tolerance (IGT) are associated with greater decline in pulmonary function. Further highlighting the importance of CFRD is the finding that CFRD is associated with increased mortality. Currently, IGT and CFRD are defined by two-hour plasma glucose (PG) during an OGTT, but PG abnormalities due to the delayed and blunted insulin secretion that are typical of CF, frequently are missed by the 2hr OGTT PG (PG2). The aim of this study is to investigate the relationship between the 1hr OGTT PG (PG1) and pulmonary function in the pediatric CF population.

Design and Methods: Retrospective chart review of all OGTT performed between 8/2005 (when annual screening was initiated) and 6/2008 in children followed by The Children's Hospital of Philadelphia CF Center. Data analysis was limited to first-time OGTT (PG0, PG1, PG2) performed in the well state. Additional data collected included PFTs (FEV1 and FVC) performed as part of routine annual evaluations at the time of the OGTT and 1-yr later, body mass index% (BMI%), sputum colonization, age, and sex. OGTT were categorized according to PG2 as normal (PG2<140 mg/dL), IGT (PG2: 140-199 mg/dL), and CFRD (PG2> 200 mg/dL) and according to PG1 as normal PG1 (PG1<140 mg/dL), IGT PG1 (PG1: 140-199 mg/dL), and CFRD PG1 (PG1> 200 mg/dL). Multivariable linear regression was used to analyze the association between FEV1 and OGTT PG. Mixed effect models were used to determine the association of OGTT PG with FEV1 one year later.Results: OGTT were available in 106 children (63M/43F; age 5.8-22 yrs); complete PG data were available in 89. In general, the CF population was healthy: baseline FEV1% (51.5+SD)=95+18% and BMI%=52+25%. By PG2 criteria, 94 OGTT were normal, 10 IGT, and 2 CFRD. Of these 94 normal, 26% had IGTPG1 and 18% had CFRDPG. PG1 was negatively associated with FEV1 even after adjustment for sputum colonization and BMI% (partial beta-coefficient=-0.1, p=0.009, R2 0.13). FEV1% was not associated with PG0, PG2, age, or sex.Conclusions: Despite being classified as normal based upon the 2-hr OGTT PG, many individuals with CF have elevated 1-hr PG. The long-term implications of having an elevated 1-hr PG for individuals with CF is not known, but the association of increasing PG1 with worse pulmonary function suggests these early PG abnormalities may have deleterious effects. Introduction: Recent advances in the treatment of cystic fibrosis (CF) have led to an increase in life expectancy; as a consequence new issues and in particular renal disease such as oxalate nephropathy, pigmented tubulopathy and CF-related nodular glomerulosclerosis in the absence of diabetes mellitus are emerging in the CF population. The importance of this is highly relevant for patients with CF given the international guidelines state a creatinine clearance <50 mg/ml/min is a contraindication to lung transplantation.

Objective: A descriptive study to determine the incidence of renal impairment in the adult CF population.Methods: The estimated Creatinine Clearance (eCrCl)was calculated using the Cockroft-Gault formula (CGF) in all adult CF patients with stable creatinine attending the Cork Adult Cystic Fibrosis Centre, and Spearman rank correlation coefficient was used to correlate this with a number of parameters of disease severity including lung function.Results: Ninety patients were recruited. Forty percent (n=36) of patients had chronic renal impairment as defined by an eCrCl <90ml/min. Despite the high incidence of chronic renal impairment only 5.5% (n=2) of the patients had a raised serum creatinine level outside the laboratory reference range, both of whom had eCrCl <50ml/min. Spearman analysis shows there is a significant correlation between declining eCrCl and declining lung function (R=0.33, p=0.002) and independently between declining eCrCl with chronic Pseudomonas aeruginosa colonization (r=0.247, p=0.02). Age, gender, CF related diabetes mellitus (CFRD), nebulised Tobramycin use and Azithromycin use did not correlate with eCrCl.Conclusion: Renal impairment is common in the adult CF population and is not confined to patients with CFRD. Serum creatinine levels alone are a very poor indicator for renal dysfunction in the CF population, identifying only 5% of patients with chronic renal impairment. This study highlights that with increasing survival and with consideration for lung transplantation there needs to be an increased consideration for regular eCrCl monitoring and caution in the use of potentially nephrotoxic medications.

McCracken, M. University of Minnesota, Minneapolis, MN, USA Cystic fibrosis (CF), a once fatal childhood disease, today has an average life expectancy of 36 years. With this dramatic improvement in life expectancy comes the challenge of preparing the adolescent CF population for adulthood. A significant piece to this is engaging these youth in self-care, which will be critical in preventing life-threatening complications and improving quality of life.Until recently, self-care has received little attention in the CF population and educational materials and tools are limited. This lack of age-appropriate materials has made working with these teens challenging. Nursing practice would be much improved if self-care materials targeted at the adolescent and young adult CF population were available.This need prompted the writing of the booklet Cruising On, Next Stop...Adulthood. Successful Strategies for Adolescents and Young Adults with CF, which provides both a process and the necessary tools for implementing self-care in the CF population.The booklet, which is evidence-based and utilizes Orem's Self-Care Deficit Theory. The booklet has four chapters focusing on helping the teen and young adult with CF develop self-care skills, learn how to integrate these shills into their daily life, and how to handle some of the challenges chronically ill teens experience as they transition toward adulthood. Several tools have been developed to help the teen successfully create and implement their self-care plan.The CF Self-Care Questionnaire measures one's knowledge of CF and its treatment plan.The Body Check is a self-assessment and decision-making tool aimed at preventing CF complications. The Self-Care Schedule puts all the pieces of the treatment plan and life activities together in a daily schedule and The Self-Care Contract helps to gradually transfer CF cares from parent to teen.Several of these tools, which were developed to improve nursing practice in implementing self-care, can also be used for nursing assessment of CF patients and nursing education in CF. They can also easily be modified for use in self-care with other chronic pediatric diseases.Digestive Care Inc. provides complimentary copies of this booklet to every CF Center in North America as a service to the CF population. Methods: We analyzed the CF Patient Registry between 1992-2007 evaluating all patients age >18 years and their level of education, employment status, and insurance status with respect to measures of severity of disease including age, sex, forced expiratory volume at one second (FEV1), and body mass index (BMI).Results: In 2007, 37.5% of patients were employed full-time, 17.3% part-time, 15.1% were disabled, and 7.9% were classified as unemployed. Although the numbers in each category have increased, these percentages have not changed since 1999. Prior to 1998, disabled patients were not categorized separate from unemployed, and the combination of the two has not changed since 1992. We found the same results for categories of level of education. In 2007, 8.6% of patients had not graduated high school, while 30.5% had graduated from college, which was no different than the percentages observed in 1995. For all categories of education and employment, there was a significant increase in median FEV1 (% predicted Cystic fibrosis (CF) patients have to spend a lot of time on inhalation every day. Patient adherence to inhaled medications is important for the outcome of clinical trials and interpretation of the results. Typically consumption of medication and diary entries are used to monitor adherence.In our study a modified investigational eFlow ® nebulizer system was used for monitoring adherence of CF patients to their inhalation treatment. For each treatment the nebulizer stores data on a chip card that needs to be inserted into the nebulizer control unit to operate the nebulizer.The chip card is replaced every time the patient visits the clinic and downloaded onto a computer using the PARI Pharma Patient Monitoring Software. This software stores all data in a database and offers several reports. Adherence is calculated as the ratio of actual to planned inhalations and is shown graphically per study day and as cumulative adherence from the start of the study onwards. This provides immediate information about the patient's adherence to the study protocol.First results within an ongoing clinical trial have shown that adherence to the study protocol varies between patients. For non-compliant patients the adherence showed peaks directly before and after they visited the clinic. The additional patient monitoring feature offers the clinical trial investigators to intervene as soon as non-adherence is recorded and to interpret clinical data accordingly. Background: Treatment regimens for cystic fibrosis (CF) are time-consuming and complex, leading to consistently low rates of adherence. To date, few studies have evaluated the use of innovative technologies to improve adherence in this population. Current infection control guidelines for patients with CF prohibit face-to-face peer contact to minimize patientto-patient transmission of potential pathogens. Thus, interventions to improve adherence must be delivered individually, limiting opportunities for peer support. This study aimed to develop and assess a web-enabled cell phone, CFFONE ® to provide CF care management information and support to improve adherence in adolescents with CF.

Methods: The acceptability, feasibility and utility of the CFFONE ® was evaluated in a focus group with health care professionals (n = 17) and interviews and surveys with adolescents with CF ages 11 to 18 years old (n = 12), adults with CF ages 21 to 36 years old (n = 6), parents of adolescents or young adults with CF (n = 12), and technology experts (n = 8). In addition, adolescents tested a prototype of the CFFONE ® (n = 9). Qualitative and quantitative data were collected from participants.Results: Focus group data with health care professionals indicated a need for this type of adherence intervention, and suggested that the CFFONE ® would be likely to improve knowledge, adherence and social support. Adolescent, adults and parents all rated the CFFONE ® as likely to improve adherence and management of the CF regimen. Technology experts rated the prototype design and format as appropriate.Conclusions: The current study provided strong support for this application of cell phone technology as an intervention to improve adherence in shown in Table 1 ). Even the patients classified as disabled had an increase in lung function with the median FEV1 reaching 40% predicted.Discussion: Although the number of adult patients is increasing, there has been no change in the general employment and educational status of patients. Overall health has improved, even in those patients receiving disability. This suggests that despite improving health status and development of adult CF programs, there has been no improvement in achievement of normal adult milestones for CF patients. (2) timeline, or views about the illness trajectory, particularly the cyclical nature of chronic disease; (3) consequences, or views about the expected outcome of the illness; and (4) control, or views on how to control symptoms. Using a validated illness perception questionnaire which we adapted for CF, we examined the relationship between illness perception and health status in a cohort of adults with CF.Methods: In the Project on Adult Care in Cystic Fibrosis (PAC-CF), a longitudinal panel study of adults with CF from 10 US CF centers, we administered a modified version of the Illness Perception Questionnaire-Revised (IPQ-R). We assessed 5 subscales from the IPQ-R: treatment control (5 items), personal control (6 items), illness consequences (6 items), illness coherence (5 items), and illness timeline-cyclical (4 items). Each IPQ-R item is scored on a Likert scale of 1-5, with a composite score summed for each subscale. Multivariate regression analyses were performed to explore the associations between IPQ-R scores and demographic (age, gender) and health status indicators (lung function, BMI, change in lung function and BMI over time, frequency of pulmonary exacerbations, and presence of Pseudomonas and B. cepacia in respiratory cultures).Results: The IPQ-R was completed by 180 respondents (response rate 73%): 63% female, mean age 36.8±10.2 years, mean FEV1 63±23% predicted. Mean (±SD) IPQ-R domain scores were: treatment control 18.1±3.0, personal control 24.6±3.3, illness consequences 23.3±3.9, illness coherence 19.6±3.9, and illness timeline-cyclical 13.4±2.7. In our regression models two common indicators of disease severity, lung function and BMI, were not associated with any IPQ-R scores. Increased frequency of pulmonary exacerbations was associated (p<.05) with increased perception of CF as cyclical (illness timeline-cyclical scale) but not with other IPQ-R subscales. Female gender was associated (p<0.05) with higher scores on both the personal and treatment control scales.Conclusions: In the PAC-CF cohort, clinical indicators of CF disease severity were not associated with illness perception. Most adults agreed that CF is a serious condition with major consequences but those with more severe disease did not see the consequences of the disease as more serious. Nor did those with more severe disease report that they have less personal control over the disease or that the treatments were less able to impact outcomes. Those who experienced a recent pulmonary exacerbation perceived their disease as more cyclical than those who had not had a recent exacerbation. Finally, age and gender were not associated with illness perceptions except that women reported higher levels of both personal and treatment control. Although HRQOL measures such as the CFQ-R have been developed for use as patient-reported outcomes, there is little data on changes in CFQ-R scores over time in adults with cystic fibrosis (CF) who are not part of a clinical trial.

Methods: The Project on Adult Care in Cystic Fibrosis (PAC-CF) is an ongoing prospective longitudinal panel study of 333 adults with CF at 10 CF centers in the US. The CFQ-R was administered to the PAC-CF cohort 7 times from April 2005 to January 2007. Individual longitudinal growth curve regression models were fit to each CFQ-R scale with clinical covariates provided by the sites.Results: Completed surveys were received from between 205 and 303 adults (response rate between 70% and 93%). Mean age of respondents at the first CFQ-R assessment was 33 years (range 19-64) and the mean FEV1 % predicted was 59.8% (SD 22%). Over 21 months of follow-up, FEV1 and number of exacerbations were significantly related to all physical domains of HRQOL, with the exception of digestive symptoms. Higher FEV1 was associated with greater HRQOL scores in all CFQ-R physical domains For example, a difference of 5% in FEV1 was associated with a difference ranging from 0.5 to 2.3 points in individual CFQ-R domain scores. Conversely, pulmonary exacerbations were associated with lower quality of life scores, as each exacerbation was associated with a decrease in CFQ-R domain scores ranging from -1.3 to -3.1 points. Weight was positively associated with scores in CFQ-R body image, eating disturbances, weight, and health perceptions domains. There were no significant population trends over time in the physical CFQ-R domain scores (physical functioning, body image, eating disturbances, digestive symptoms, respiratory symptoms, weight, health perceptions, vitality). However, there were longitudinal population time trends in 3 psychosocial CFQ-R domains: treatment burden (+8.9 points per year), emotional functioning (+3.2 points per year), and social functioning (-2.4 points per year). Individual variation in slopes was seen over 21 months in both physical and psychosocial subscales of the CRQ-R. For example, for 5%-95% of the population (excluding outliers), the range of linear slopes was -3.01 to +3.82 points per year for the respiratory subscale, -6.44 to +8.92 points per year for the weight subscale, and -5.54 to +2.83 points per year for the role subscale.Conclusion: In a population of adults with CF who are not part of a clinical trial, clinical variables such as FEV1, exacerbation frequency, and weight were correlated with scores on the matching subscales of the CFQ-R. For the population examined as a whole, psychosocial domains of HRQOL such as treatment burden and social functioning change over 21 months, but the physical domains such as respiratory functioning are stable. Despite this lack of trends over time in the population, there was substantial variation in the individual-level slopes in the physical domains of the CFQ-R over the 21 month period.

Eisenstadt, I.; Cohen, T.; Wexler, I.; Kerem, E. CF Center, Hadassah Hospital, Jerusalem, Israel Background: Cross-sectional studies have shown that pregnancy/normal term delivery is safe for women with cystic fibrosis (CF), but may be a risk factor for women with advanced lung disease. The potential hazards may trigger a moral conflict between the caregiver's advice against pregnancy and the patient's desire for motherhood.Aim: 1. To compare the views of caregivers and women with CF with respect to pregnancy. 2. To analyze the relationship between the severity of the disease and the decision to become pregnant.Methods Objectives: Patient-reported outcomes (PROs) are increasingly being used in clinical trials and clinical care as health outcomes. Although assessment of health-related quality of life (HRQOL) in young children is challenging, they have a unique and important perspective on how cystic fibrosis (CF) affects their symptoms and daily functioning (e.g., school/daycare). HRQOL has been reliably evaluated in children with CF ages 6 and older using the Cystic Fibrosis Questionnaire-Revised (CFQ-R). The CFQ-R is one of the most widely used HRQOL instruments for CF and was classified as "well-established" in a review of HRQOL measures (Palermo et al. J Pediatr Psychol 2008; 33:983) . The purpose of this study was to downwardly extend the CFQ-R for children 3-6 years of age using a pictorial format.Methods: The CFQ-R preschool version was adapted from the CFQ-R child (ages 6-13) version. Pictures were used to facilitate preschooler's responses, and items were modified to include appropriate developmental examples (see Figure 1 ). Following completion of the initial draft of the CFQ-R preschool version, psychometric properties were evaluated. Pediatric patients (ages 3-6) and their parents completed the CFQ-R and Ped-sQL, a generic measure of HRQOL. Background and medical information were also obtained.Results: Fifty-five pediatric patients (M = 4.89, SD = .88) and their parents in Australia participated. On average, children had mild lung disease as measured by FEV 1 % predicted (M = 93.46, SD = 17.12). Internal consistency (Cronbach α's) ranged from .60 to .68, with the exception of the Eating Disturbances, Treatment Burden, and Social Functioning scales. For the parent version, internal consistency was adequate to good (α = .66 -.88), with the exception of the Health Perceptions, School and Emotional Functioning, and Vitality scales. These results suggest that some scales, such as Social Functioning, may not be appropriate for preschoolers and some questions need to be re-worded to improve the instrument's psychometric properties. Significant correlations were found between relevant CFQ-R and PedsQL scales for both the children (r's = .36 -.46) and parents (r's = .29 -.78), providing evidence of convergent validity. Good agreement was also found between children and parents on the Respiratory Symptoms scale (r =.44, p<.001).Conclusions: Preliminary evidence of reliability and validity of the CFQ-R preschool version was found; however, revisions to the preschool CFQ-R are needed, as well as further validation.Supported by NHMRC and RCHF Brisbane. incorporated demography, disease severity, personal opinion on pregnancy, open questions on anxiety during/after pregnancy. Results: Thirty-six patients (70% of all adult females with CF, mean age 27.5 yrs, FEV 1 57.5%, PI 71%, CFRD 41%) and 36 caregivers (average age 43.3 yrs, 11 male/25 female) from all CF centers in Israel responded to the questionnaire. Sixteen CF patients gave birth to 32 children. Caregivers, more than patients, perceived severity of disease to be the limiting factor in pregnancy. Both caregivers and patients agreed that high pulmonary functions (FEV 1 м90%) are no obstacle to pregnancy, for moderate pulmonary functions (FEV 1 Ϲ50%) caregivers more than patients advise to refrain from pregnancies. In contrast to patients, caregivers prefer the wellbeing of the patient over the health of the fetus. A relationship between religious observance of the patient and objection to abort were noted. Treatment compliance during pregnancy was higher than predicted by caregivers.Conclusions: A gap in the perception of pregnancy was observed between caregivers and CF patients. Pregnancy counseling should therefore be incorporated into the treatment plans for adult women at CF centers. Objectives: Numerous studies have demonstrated that patients with chronic illnesses are at increased risk for anxiety and depression; however, no systematic evaluation of symptoms of anxiety and depression in patients with cystic fibrosis (CF) has been conducted (Quittner et al., 2008) . This study estimated the prevalence of anxiety and depression in patients with CF and parent caregivers and examined their relationship to health outcomes, such as pulmonary functioning and health-related quality of life (HRQOL).Methods: During a routine clinic visit, adult patients with CF and parent caregivers of children with CF completed two screening measures of depression and anxiety: the Hospital Anxiety and Depression Scale (HADS) and the Center for Epidemiological Studies -Depression (CES-D). Participants also completed a demographic and medical information form, which was confirmed by chart review. Adult patients also completed a disease-specific measure of HRQOL, the Cystic Fibrosis Questionnaire-Revised (CFQ-R).Results: Forty-five adult patients (M = 30.51 years, SD = 9.18) and 11 parents (M = 40.0 years, SD = 7.65) participated. On average, adult patients had moderate lung disease as measured by FEV 1 percent predicted (M = 55.01, SD = 20.25). Furthermore, 15% of adult patients and 18% of parent caregivers were currently taking anti-depressant or anti-anxiety medications. In the adult sample, 33% endorsed elevated anxiety symptoms and 11% endorsed elevated depressive symptoms on the HADS. Adults reported slightly more symptoms of depression on the CES-D (16%) than the HADS (11%). Nine percent reported co-morbid depression and anxiety. Among parents, 46% endorsed elevated symptoms of anxiety, 27% endorsed elevated depressive symptoms on the HADS and CES-D, and 27% reported comorbid symptoms. Elevations in anxiety or depression were related to worse HRQOL. Specifically, anxiety was related to worse Role and Emotional Functioning (r's = -.33 to -.54). On the HADS, depressive symptoms were negatively related to Emotional Functioning, Body Image, Eating Disturbances, and Health Perceptions (r's = -.30 to -.42). Elevated scores on the CES-D were negatively related to a majority of CFQ-R domains (r's = -.30 to -.54). Patients currently on anti-depressant or anti-anxiety medications also reported worse Role Functioning (r = -.38).Conclusions: In general, higher rates of anxiety versus depression were reported by adult patients and caregivers. Both symptoms of anxiety and/or depression were related to worse HRQOL. Due to clinically elevated levels of anxiety and depression in CF patients and caregivers, routine screening during annual clinic visits is recommended. The current international prevalence study (tides-cf.org) will provide additional data.Funding provided by the Cystic Fibrosis Foundation. Musculoskeletal pain is recognised as a complication in cystic fibrosis (CF) but the prevalence and clinical significance of this co-morbidity in adults with stable lung disease has not been well described. The aim of this study was to determine the location, intensity and experience of musculoskeletal pain and its impact on quality of life in individuals with CF. Participants with clinically stable CF from an outpatient clinic completed three questionnaires measuring the location, severity and degree of interference in daily activities of musculoskeletal pain (Brief Pain Inventory), the psychological impact of pain (Pain Catastrophising Scale) and quality of life (Cystic Fibrosis Quality of Life questionnaire). Seventy-seven participants with mean age 29 years (SD 8 years), FEV1 60 (25) % predicted were included. The prevalence of musculoskeletal pain was 89% and was not influenced by lung disease severity. Pain was commonly located in the back, buttocks and hips (70%), head and neck (52%) and lower limbs (29%). Although pain intensity was generally mild, participants with musculoskeletal pain reported poorer physical functioning (p=0.01) and increased interference with treatment aspects of CF (p=0.03) compared to those with no pain. Participants with symptoms of CF arthropathy reported greater intensity and impact of pain. Pain catastrophising and FEV1 were independent predictors of quality of life in adults with CF. Musculoskeletal pain is a significant complication which negatively influences quality of life in adults with stable CF. These results lend support to the inclusion of musculoskeletal pain screening as part of CF management. As the expected age of survival has increased so has the opportunity to meet major life milestones for adults with cystic fibrosis (CF). Recently, at the Adult Cystic Fibrosis Center in Morristown, the husband of a 37-year-old female patient had a stroke. Thankfully, he has fully recovered. However, long term financial planning is a new frontier for the CF care team. It cannot be assumed that our patients will die before their spouse. Patient education in regards to insurance needs has been focused on health insurance, but perhaps should be expanded to include life insurance.

Methodology: This retrospective chart review included married patients who had been asked about the possession of life insurance for themselves and/or their spouse. Information obtained included: 1) gender, 2) age, 3) employment status, 4) life insurance of spouse, 5) life insurance of patient.Results: Of the 80 patients presently attending our adult CF center, 34 of them are married (42.5%). Upon chart review, 15 (44%) of the patients who are married had information regarding their life insurance status and that of their spouse. The females numbered 11 (73%) and males 4 (27%). Age range of these patients is 21-59 years with a mean age of 38.7 years. Ten patients (67%) are employed full or part-time. The number of spouses of our married adult patients who have life insurance are 13 (87%), with 11 (73%) of our patients owning some life insurance policy. Three (20%) patients were late-diagnosed adults and had purchased additional life insurance prior to their diagnosis.Limitations: This is a small sample from only one center. Documentation was not complete on amount of life insurance coverage. Patients and spouses may be underinsured. Focus was only on married adults, single adults should also be evaluated in the future. Discussion: As the expected age of survival increases, our adult patients move through different adult milestones. The CF healthcare team should expand the area in which we educate our patients, in preparation for these milestones. It can no longer be assumed that our married patients will not be surviving partners. The healthcare team can be proactive in guiding and educating our married patients to pursue appropriate financial and estate planning with their spouses. All adult CF patients need to understand that they may be eligible for life insurance provided by their employer and/or union and be encouraged to take advantage of that benefit as soon as possible. Greater awareness and sensitivity on the part of practitioners working with CF patients is needed to educate and prepare proactively, rather than providing aid to families in a difficult period like an untimely death. Results: In total, 865 adults with CF (M = 30.6 years, SD = 8.5; 55% male) completed the survey. The sample was predominantly Caucasian (95.2%; 2.2% Hispanic, 1.6% African American, 1% Other). Most participants reported being diagnosed at ≤ 1 year of age (60.2%, M = 4.8 years, SD = 8.3, range = 1-67 years). More than 50% of participants completed some college, and more than one-third were working full-time.

Participants reported taking responsibility for their health care and attending clinic visits alone during adolescence (M = 16.5 years, SD = 5.5; M = 18.2, SD = 11.0, respectively). Average age at full transition to adult care was 22.0 years old (SD = 6.6). Twenty-three percent of participants had not transitioned at the time of the survey. On average, participants felt "prepared" at the time of transition (M = 2.9, SD = 0.9). The age at which patients took responsibility for their health care was associated with the age at which they attended clinic visits alone (r = .24, p < .001) and transitioned to adult care (r = .32, p < .001). Barriers to transition included satisfaction with their current CF care team, convenient access to the clinic, and reservations about transitioning to a new CF care team. Advice from adult to younger CF patients included being adherent to their treatment regimen, seeking information, and establishing trust with medical providers. Content analysis will be conducted to identify common themes related to barriers to transition and advice to younger patients.Conclusions: Adults with CF reported preparing for transition during adolescence, which included taking more responsibility for treatments, attending clinic visits alone, and transitioning to adult care. The percentage of adult patients seen by adult providers and the barriers to transition reported in this study were consistent with previous literature. Limitations of this study include the use of retrospective data from participants and lack of access to their health data. Future research should assess the health care team's perceptions of preparedness and examine predictors of successful transition.

This study was supported by Novartis Pharmaceuticals Corporation.questions or concerns. Special exceptions may be made for patients with developmental disabilities or psychosocial concerns. Upon completion of solo clinic visits and all homework assignments, patients attend a formal Adult Transition Clinic, in which they are seen by both the pediatric and adult care teams. This clinic is held at our pediatric care center and serves as the final step in the Journey to Independence program. Results: Since January 2008, a total of seven patients have transitioned to adult care. Seven additional patients are scheduled to transition in September 2009. One patient was enrolled and had completed over half the program, when she transferred to another pediatric care center. Of the seven patients scheduled to transition this year, six have had at least one solo clinic visit and have completed the My Med Sheet and My Meal Plan homework assignments; five patients have completed the Who You Gonna Call? assignment; one patient has completed the Challenges of Independence assignment. Three additional patients have started the program, had a solo clinic visit, and completed one homework assignment. Three other patients eligible by age have not participated because of special exceptions mentioned previously.Conclusion: Journey to Independence has been well received by patients, parents, and staff, and we have met our initial objectives. Goals to further develop the program include: Create additional materials to help patients achieve educational objectives; enhance the Scout, Explorer, Trail Guide, and Deputy Stages; develop artwork to establish a "look" for the program, including a permanent visual aid in clinic so patients can follow their progress on the journey; and measure patient and parent satisfaction with the program.

Butcher, J.L.; Nasr, S.Z. Pediatrics, University of Michigan, Ann Arbor, MI, USA Objective: Research has shown that rates of medical regimen adherence for children diagnosed with cystic fibrosis (CF) are less than 50% and that adherence tends to decline during adolescence. To address this concern, researchers have begun to develop intervention programs designed to promote adherence. One promising intervention, developed by Alexandra Quittner, utilizes a problem solving approach between CF patients and their families. Based on this work, the "CF My Way" program was created by an interdisciplinary team of CF clinicians to meet the need for an intervention that can be implemented within a CF Center. The objective of this pilot project was to examine the clinical utility of "CF My Way."Methods: Four CF Centers across the United States participated in the pilot of "CF My Way." Data are reported from 12 patients from our center and 13 patients from across all four centers. Patients ranged in age from 14 to 20. Adolescent patients were asked to participate in the pilot program during a routine clinic appointment or an inpatient hospitalization. Participation included completing an electronic knowledge assessment, reporting rates of adherence, identifying barriers to adherence, and engaging in a problem solving session with a family member. Following participation, patients had the option to provide feedback on the program. Results: For the 12 patients from our center, the average score on the knowledge assessment was 90% with highest average scores in the Lung Health section (92%) and lowest scores in the Nutrition section (84%). All patients identified at least one area of their medical regimen where they were less than 100% adherent (average number of problem areas=3.75, range=1-12) with 50% choosing to focus on adherence to nebulized medications. The most common barriers identified included forgetting, lack of time, and low motivation to complete treatments. Each of the 12 patients was able to choose a solution with a family member with 40% choosing a solution involving obtaining an incentive/reinforcement for adherence, 35% choosing a solution involving use of a reminder, and 25% choosing a solution to increase the ease or decrease the time associated with treatment completion. Themes from 13 patients across the four sites who provided feedback on the program suggest that patients thought that the program increased their understanding of CF, helped them to realize that their adherence was poor, and identified ways to overcome barriers to adherence. However, one to two weeks after the session, only half of the participants reported that they were implementing the solution generated. Parental support was identified as a key factor related to patients continuing to implement their solution. Background: Transition of young adults with cystic fibrosis (CF) from pediatric to adult care programs is an important priority as the majority of individuals with CF are living well into their fourth decade and almost 45% are over 18 years old.

Objectives: (1) To estimate the prevalence and timing of transfer from pediatric to adult CF care centers for adolescent and young adult patients over the age of 18 years old; and (2) To describe and compare young adults between the ages of 18-25 years cared for in pediatric versus adult CF care programs with regard to demographics, pulmonary function, and BMI.Methods: A multicenter retrospective cohort study using the U.S. National CF Foundation Patient Registry. Transition cases were identified using CF Program ID change from pediatric to adult program for all 18-25 year olds who transferred between 2003 through 2007, and who had at least three years of consecutive data (one year prior, the year of, and one year following transfer). Descriptive statistics were performed. Data from subjects cared for in combined pediatric/adult CF Programs were excluded from this analysis.Results: There were 406 patients, aged 18-25, who had a single change from a Pediatric to Adult CF Program ID between the years 2003-2007, which is 9.4 percent of all "eligible" patients (406/4331). The mean age of patients in their year of transfer was 21.2 ± 1.3 years. Cases who transitioned care were less likely to be female (49% vs 54%, p=0.03), were younger (21.2 vs 22.3, p<0.001), had on average more outpatient visits per year (4.5 vs 4.0, p=0.001), higher mean number of respiratory cultures per year (2.8 vs 2.4, p<0.001) and higher FEV1 % predicted (67 vs 65, p=0.05) than patients who did not transfer to an Adult CF Program. There were no differences in mean BMI % predicted, mean hospitalizations per year, or mean FVC % predicted between the two groups.Conclusions: Patients who were transferred to an Adult CF Program were more often male, younger, had higher mean FEV1 % predicted, and seen more often in the outpatient setting than those that remained in pediatric care. Further research is needed to explore the relationship between CF transition practices and health status outcomes in order to establish a systematic, evidence-based transition process.

Shipp, A.; Siragusa, A.; Scott, P.; Crews, B. Children's Healthcare of Atlanta, Atlanta, GA, USA Background: In 2007, our pediatric cystic fibrosis (CF) care center recognized the need for a formal transition process and subsequently developed a program called Journey to Independence. Program objectives include: Assess patients' transition readiness; offer opportunities for patients to discuss concerns and ask questions independent of parent input; establish educational goals and create homework assignments in order for patients to meet those goals; and increase involvement with the adult care team.Methods: The Journey to Independence program is composed of five stages: Scout (ages 8-10); Explorer (ages 11-13); Trail Guide (ages 14-15); Deputy (age 16); and Park Ranger (ages 17-18). Each stage includes five components with specific educational objectives: Understanding CF; Respiratory; Nutrition; Treatment; and Lifestyle. To obtain the objectives in each stage, patients are required to complete various assignments. A patient in the Park Ranger Stage is expected to: Describe his daily medication and treatment regimen (My Med Sheet); create a weekly food diary (My Meal Plan); determine who to call with certain health related questions (Who You Gonna Call?); and identify and describe how to meet potential challenges of selfcare (Challenges to Independence). In addition, patients in the Deputy Stage are required to attend their clinic appointments without parental accompaniment (solo clinic visits). Parents are asked to remain in the waiting room until the end of the visit, when they are invited to the room to address their Conclusions: This pilot project supports the clinical utility of "CF My Way." Adolescent patients described the program as informative and valuable, were able to complete the program components within a medical setting, and effectively worked with a family member to develop an acceptable solution. However, only half of adolescents actually implemented the solution. Future trials of this program should emphasize parental support and involve follow up with the adolescents after participation.Supported by Novartis Pharmaceuticals Corporation. To determine if attitudes can be enhanced and adherence to treatment regimen improved in adolescents with CF by using a simulation game.

Methods: In 2001, the game "My Life" was developed and proven effective. The game was designed to improve decision making and problem solving skills of adolescents. Game participants live out the life of a fictional character. Participants decide how to assign available "effort points" (or time) among a variety of "possessions" (or activities). How participants spend their effort points will influence and shape their character's life. The game is designed to be statistically similar to chances in real life. Post game discussions reinforce lessons learned in the game and link the events and decisions to the participant's own life. "My Life with CF" was adapted from the original "My Life" game except the fictional character has CF. Participants also make decisions regarding nutrition, treatments, and medications. The game incorporates realistic projections of the effects of the disease and adherence to treatment on the life of the fictional characters with CF.Sixteen participants with CF, between the ages of 12-20, were enrolled in the study. They completed 4 Attitude Assessment Scales before the game. Attitude scales were completed again immediately after playing the game and one month later. Seven of the participants returned the forms at one month. Attitude Scales 1, 2 and 3 were previously validated. Attitude scale 4 was created specifically for the "My Life with CF" game. Each attitude scale has 20 questions and is based on a Likert-type scale. Scale 1 assesses personal invulnerability Scale 2 assesses present versus future considerations Scale 3 assesses luck versus effort Scale 4 has an equal number of questions from the above three categories, specific to CF therapy.Results: Changes in scale scores over time are summarized in the scale below:Participants' feedback: I learned how decisions you make early on can have an effect on your life later on.I stay up as long as I have to to finish my treatments.The game really made me think about how I want to live. Conclusions: Attitudes were enhanced by using a simulation game "My Life with CF". Attitudes toward CF therapy remained enhanced one month after playing the game.

Zubrinich, C.M. 1 ; Gilljam, M. 2 ; Chowdhury, N. 3 ; Singer, L.G. 3, 4 ; Tullis, E. 1 1. Department of Respirology, St Michael's Hospital, Toronto, ON, Canada; 2. Department of Pulmonary Medicine and Allergology, Sahlgrenska University Hospital, Goteborg, Sweden; 3. Toronto General Research Institute, University Health Network, Toronto, ON, Canada; 4. Faculty of Medicine, University of Toronto, Toronto, ON, Canada Background: Quality of life (QOL) is increasingly recognized as an essential component of assessing new therapies and individualizing patient care, so much so that the FDA now mandates the inclusion of a QOL measure in evaluations of new therapies. Cystic fibrosis (CF) imposes unique effects upon the QOL of those affected as there may be multiple organ systems involved, treatment may be complex, time-consuming and require conscientious effort, and end-of-life decisions may be faced many years before an individual's peers.In earlier work (Gilljam M. Pediatr Pulmonol 2002;S24:348), a CF-specific QOL instrument was developed with input from multiple individuals with CF and their medical caregivers from three countries. Using frequency-importance ranking for item-reduction, 150 potential questions were distilled to the 40 most important, thus creating the CF-QUEST. This process was done in English and Swedish.The aims of this study were to test the validity and responsiveness to clinical change of the CF-QUEST by comparing it to other QOL measures, physical measures of health, complexity of disease, treatment burden and response to therapy for an acute infective exacerbation.Methods: The CF-QUEST was administered to adults with CF who were deemed to be clinically stable, and to a second group who were experiencing an infective exacerbation of pulmonary disease. This second group completed all assessments on two occasions, once at the start of the exacerbation and then again when they had completed treatment and returned to baseline; clinically stable patients completed all measures once only. Other existing QOL instruments were concurrently administered: a general health QOL questionnaire (SF-36), a respiratory QOL questionnaire (St George's Respiratory Questionnaire), a CF specific QOL questionnaire (the CFQ-R Adult) and a Visual Analogue scale. Measures of physical health that were assessed for correlation included pulmonary function tests, vital signs, sputum pathogens, body mass index (BMI), CF-related co-morbidities and daily treatment time.Results: Over the initial ten weeks, the CF-QUEST has been administered to 58 adults with CF; recruitment is ongoing. Forty men and eighteen women have been included, with ages ranging between 18 to 63 years. The FEV1 of the participants has ranged between 22 to 103 percent predicted.An interim analysis of the data currently available has revealed that there is a significant correlation between the CF-QUEST score and other QOL tools (Spearman r = 0.84 for Mental health component of SF-36). With a target of 100 adults, we anticipate recruitment will be complete by the end of July 2009.

The CF-QUEST is a QOL tool developed specifically for use in adults with CF. Field testing to establish validity and responsiveness is necessary before this measure can be used in trials of new therapies.

impact of psychiatric disorders on CF care was identified in 14 (78%) cases. Psychosocial or educational interventions were implemented in all 18 cases. Psychopharmacologic treatment of ADHD was attempted in 13 (72%) of cases.Psychopharmacologic "Best Regimens" included stimulants in 5 cases, with ADHD monotherapy in 9 cases and a combination of 2 ADHD medications in 3 cases. A Clinical Global Improvement (CGI) rating of Much Improved or Very Much Improved was achieved in 8 (62%) of the 13 cases in which ADHD medications were tried. These well tolerated and effective regimens included atomoxetine; atomoxetine + guanfacine; mixed amphetamine salts; buproprion XL + OROS-methylphenidate; methylphenidate; nortriptyline; nortriptyline + guanfacine.While ADHD medication trials were not generally associated with clinically significant BMI changes, a significant drop in BMI occurred with some stimulant trials. Nortriptyline use was associated with a robust increase in BMI percentile, from 22% to 61% in one case and from 34% to 76% in another case.Conclusions: ADHD is common and treatable in pediatric CF patients and often occurs in conjunction with other psychiatric conditions. Non-stimulant medications are frequently effective in this population and may be better tolerated than stimulants in patients at risk for weight loss. Further investigation of the effects of ADHD on medical treatment adherence in CF and of psychosocial and psychopharmacologic treatments for ADHD in the CF population is warranted. 

Catastini, P.; Picchi, M.; Lattughi, E.; Repetto, T. Pediatric, Meyer Hospital, Florence, Italy Introduction: Cystic fibrosis (CF), as a multi-organ disease affecting nutrient absorption and assimilation, often impacts the ability of patients to achieve and maintain adequate body weight. This aspect of the illness could in turn render a patient weaker and less ready to confront an exacerbation of their condition. It is therefore frequently recommended that patients maintain a high calorie diet and sufficient body weight, especially during moments when the effects of the illness are most severe. For this reason, body image and the relationship with the body in general is often stressed in patients with CF.Objectives: The study seeks to determine if there is a correlation between levels of anxiety and depression and body image in CF patients.Method: The Hospital Anxiety and Depression Scale (HADS) questionnaire was administered to 69 patients, 37 females of mean age 27.63y (±6.66), BMI 20.92 (±3.24) and 32 males of mean age 29.86y (±10.15), BMI 22.32 (±4.05) during routine clinical checks.Result: A significant difference in the body image of male and female CF patients was found.The BMI of male patients negatively correlated with levels of anxiety (r=-.386,p<.05), while the BMI of females positively correlated with both levels of anxiety (r=.377,p<.05) and depression (r=.406,p< .05).Conclusion: This study discovered differences in the emotional states of males and females CF patients with a normal BMI, the latter group showing higher levels of anxiety and depression as BMI increased. It is possible that this mirrors a difference which can be found in the healthy population, where a strong body reinforces male identify, while an increase in body weight has a negative impact on female self-image.With regards to females, this study demonstrates that CF patients have the same mental constructs of body image as their healthy peers and seek to achieve the same "slim" body image. However, this could represent a risk for this group as it may impact adherence to a high calorie diet, enzyme therapy and supplementation, presenting the possible need for psychological intervention for correct clinical management.Supported by: Italian CF Foundation

Methods: Twenty-five patients (ages 7 to 18 years) completed the Memorial Symptom Assessment Scale (MSAS), the HADS and CES-DC for depression/anxiety, and the CFQ-R for QOL. A score of less than 70% on a CFQ-R domain (10-20% below the median on any domain) was designated as a "CFQ problem score." MSAS item and domain scores >1 were considered an "MSAS problem score" based on published validation statistics. Additional questions assessed abdomen, chest, limb, and joint pain, consistency with treatments, and presence of pancreatic insufficiency. Analysis methods included the non-parametric Mann Whitney U to compare medians and chi-square or Fisher's Exact test for dichotomous variables. Pearson correlation coefficients were evaluated. Exact statistical methods were used to adjust for the small sample size.Results: Body Image: Eight of 25 (40%) of responders had "CFQ problem scores." CFQ-R Social (p<.001) and Emotional (p=.03) domain scores were significantly lower for those with body image "CFQ problem scores." MSAS Physical domain scores were significantly higher for those with body image "CFQ problem scores" (p=.04). Patients with pancreatic insufficiency were more likely to have a problem score (56.3%) compared to those not having a problem score (11.1%) (p=.04). Pain: Eight of 25 (32%) patients indicated having frequent pain. Forty-eight percent and 44% experienced pain in their limbs and joints respectively. Of respondents who indicated frequent pain 62.5% were severely depressed (CES-DC). The MSAS pain score highly correlated with MSAS Lack of Energy (r=.57), Drowsy (r=.7) and Worry scores (r=.59). Sleep: Five of 24 (20.8%) patients had MSAS Sleep domain scores greater than 1. Sleep scores highly correlated with MSAS Lack of Energy (r=.5) and Drowsy (r=.63). MSAS sleep scores tended to be higher for patients with severe depression (CES-DC) (p=.07).Depression/Anxiety: Eight of 13 (61.5%) patients completing the CES-DC were severely depressed. Four of 13 (30.8%) on the HADS were highly anxious and 2 were mildly depressed. Patients positive on the HADS were positive on the CES-DC. MSAS Psychological domain scores were significantly higher for depressed patients (p=.01).Conclusion: Results suggest that pediatric patients with CF have a negative perception of their bodies, experience significant sleep problems, have limb and joint pain, and suffer from severe depression/anxiety. Based on these results, we will continue data collection to further examine the impact of these symptoms on QOL. Our ultimate goal is to improve the lives of patients with CF through a better understanding of their physical, emotional, and social functioning. The results of this study will aid in the development, implementation, and evaluation of future interventions. Methods: Retrospective chart review of patients aged 5-18 in the MGH Cystic Fibrosis Program referred from 8/05-12/08 for outpatient child psychiatric consultation and diagnosed with ADHD using DSM-IV-TR criteria. Treatment outcomes were assessed from recorded clinical data including MGH ADHD Rating Scales prospectively completed by the child psychiatrist, parents, and/or teachers as well as retrospective clinician determinations of severity and response following interventions. For each patient who underwent ADHD medication trials, the "Best Regimen" resulting in greatest improvement in ADHD symptoms with the least clinically significant side effects was identified.

Results: Of the 188 patients aged 5-18 treated in the MGH Cystic Fibrosis Program during the study period, 18 (9.6%) were referred by the CF team for child psychiatric consultation and diagnosed with ADHD, comparable to ADHD prevalence rates in the general population (1) . Additional psychiatric diagnoses were present in 17 (94%) of the 18 CF patients diagnosed with ADHD, including mood disorders in 12; anxiety disorders in 5; adjustment disorders in 4; and autism spectrum disorders in 2. An adverse Catastini, P.; Bertini, G.; Martellacci, A. Pediatrics Dept., Meyer Hospital, Florence, Italy Introduction. Cystic fibrosis (CF) is characterized by progressive decline in lung function up to the end-stage of respiratory failure. The impact of CF on pulmonary function might affect the ability to obtain refreshing sleep. Sleep problems often remain undiagnosed though they have a deep effect on people: daytime somnolence, headache, difficulty concentrating.

Objectives. The objective of this study is to determine the quality of sleep as it is perceived both in CF patients and in an age and sex homogeneous healthy control group. Furthermore, the study will investigate the causes and the impact of sleep problems.Method. Only CF patients with FEV1 >50% were included in this study. A self-report sleep questionnaire made up of 17 items was administered to 27 adults with CF: 15 females of mean age 24.96 y (±10.51), and 12 males of mean age 21.35 y (±8.94). The same test was administered to 27 adults of a control group: 15 females of mean age 22.33 y (±6.77), and 12 males of mean age 25.78 y (±7.43).A self-report sleep questionnaire with 19 items was also administered to 10 parents of children with CF, 5 females and 5 males, of mean age 9.79 y (±1.79). The same test was administered to 10 parents of healthy children, 5 females and 5 males, of mean age 10.63 y (±2.97).Results. Of the CF patients who responded, 21.6% experienced sleep problems while only 10.8% of control group who responded reported problems (t-test, p=.045). Differences between CF patients and the control group have emerged. CF patients reported difficulty with dry mouth upon awakening due to open-mouth breathing (t-test, p=.027 and p=.016). In the CF patients FEV1 is negatively correlated with facility in falling asleep (r=-.331, p<.05). There are significant correlations between CF adults and healthy adults about the modality of sleep (r=.296, p<.05), the facility of awakening (r=.475, p<.05), the feeling of awakening (p=.376, p<.001) and the quality of their sleep (r=.315, p<.05).Conclusion. The results of the present investigation have revealed that sleep problems appear to be significant in people with CF. In CF subjects, high levels of FEV1 are accompanied by a greater facility to fall asleep. Compared to the control group, the CF group has shown greater difficulties in falling asleep and this is the cause of a greater fatigue and sleepiness when awakening. For these reasons the quality of sleep perceived is worse in CF patients than in the control group. These results suggest that the CF patients show meaningful levels of sleepiness in the morning.How much could this be correlated with the patient's Quality of Life? 

Background: Numerous studies have demonstrated that individuals with cystic fibrosis are at increased risk for developing psychological difficulties such as anxiety and depression. As these psychological outcomes can vary considerably across affected individuals, it is necessary to identify which factors are associated with these individual differences in outcome. A potentially protective factor in these individuals' psychological functioning is acceptance.Objective: This study aims to prospectively investigate the role of acceptance in anxiety and depression in adolescents with cystic fibrosis. Thereby, this study replicates and extends existing research that has heavily relied on cross-sectional methods. We expected that higher levels of acceptance would be related to lower levels of anxiety and depression.Methods: Forty adolescents with cystic fibrosis (14-22 years) completed the Illness Cognition Questionnaire and the Hospital Anxiety and Depression Inventory (Time 1). After 6 months, 28 of these adolescents again completed the Hospital Anxiety and Depression Inventory (Time 2).

A set of hierarchical regression analyses was performed to examine the role of acceptance in the psychological functioning of these adolescents. Acceptance at Time 1 was related to both anxiety (β=-.49, p<.05) and depression (β=-.41, p<.05) at Time 1. Acceptance at Time 1 was also related to depression at Time 2 (β=-.49, p<.05).Conclusions: As expected, higher levels of acceptance were related to lower levels of anxiety (Time 1) and depression (Time 1 and 2) . These results suggest that acceptance (a) should be taken into account in the multidisciplinary treatment of those individuals with cystic fibrosis who exhibit psychological difficulties, and (b) can be an important factor in the prevention of depression. Furthermore, addressing depression might also have indirect positive consequences on health outcomes, as depression is related to poorer treatment adherence and disease management.Acknowledgements Objective: Youths with cystic fibrosis (CF) may be at increased risk for internalizing symptoms such as anxiety and depression, but little research has examined risk factors that may increase risk for developing such problems among youth with CF. Monitoring, a cognitive-affective information processing style characterized by a tendency to scan for, attend to, and magnify threatening cues when stressed, was recently found to be a concurrent correlate of increased internalizing symptoms (Bennett et al., 2008) . The purpose of the present study was to examine whether monitoring and blunting, which is characterized by the use of distraction or ignoring when stressed, predict increases in future internalizing symptoms.Methods: Participants (ages 7 to 23; current n=43) attending a regularly scheduled CF clinic visit completed the Children's Behavior Style Scale (Miller et al., 1995) to assess monitoring and blunting at two time points separated by 2 years. Participants and a parent also completed measures of participants' internalizing symptoms (Children's Depression Inventory-Short form [CDI-S], Kovacs, 1992; Pediatric Symptom Checklist, Jellinek, Murphy, & Burns, 1986 ; State-Trait Anxiety Inventory for Children [STAIC], Spielberger, 1973) at both time points.Results: Youth who were relatively high on monitoring and low on blunting (i.e., high monitoring -blunting scores) had increased internalizing symptoms at time 2 on each of the three internalizing measures. Controlling for time 1 depressive symptoms (CDI-S), monitoring-blunting at time 1 predicted greater depressive symptoms at time 2 (r = .31, p < .05). Similarly, controlling for time 1 trait anxiety (STAIC), monitoring-blunting at time 1 predicted greater trait anxiety symptoms at time 2 (r = .32, p < .05). Finally, controlling for time 1 internalizing symptoms (PSC), monitoring-blunting at time 1 predicted increased parent-rated internalizing symptoms at time 2 (r = .39, p < .05).Conclusion: An affective information-processing style characterized by relatively high monitoring and low use of blunting appears to place youths with CF at increased risk of future internalizing symptoms.The objective interpretation of the results of the CFQ is restricted by the lack of reference data from a healthy population. The aim of this study was to collect these reference data.Subjects and methods: A cross-sectional study was done. A multistage, random sample was taken from all primary and secondary schools in the city of Nijmegen. The CFQ was completed by 491 healthy children and adolescents (age groups: 6-11, 12-13 and ≥ 14 yrs). Children with a chronic disease or visiting a paediatrician were excluded. Specific disease-related questions were omitted in the results.Results: In this population the average score was between 80-90 per domain on a scale from 0-100. Lower scores were reported in the domains Emotional State in all age groups and Vitality in the age group ≥ 14 yrs. The domain Body Image scored significantly lower in the age group ≥ 14 yrs (p < 0.01). Boys ≥ 14 yrs showed higher scores in the domains Emotional State, Physical, Eating and Health Perception compared to girls. No relationships were found between education or socioeconomic status and CFQ scores. The age group ≥ 14 yrs showed consistent results in all domains (Cronbach's alpha ≥ 0.6). In the age groups 6-11 and 12-13 yrs the CFQ consistency was slightly lower in some domains (Cronbach's alpha 0.3-0.8).Conclusion: Mean CFQ scores in the healthy population are not constant across the ages; this should be considered when interpreting the CFQ scores in CF patients. Trends in generic domains can not only be explained by having CF as there is also a trend visible in the different healthy age groups. These CFQ reference data are unique and provide objective criteria to define abnormal versus normal CFQ scores in CF patients. Objective: Interest in exercise has increased recently because of its health benefits for patients with cystic fibrosis (CF). In randomized controlled trials, exercise improved lung function, muscle strength, and quality of life (1) . There is also evidence that patients with good aerobic fitness have better survival than those with lower levels of fitness (2) . More CF care teams have been recommending exercise to their patients, with 24.6% of CF centers now prescribing exercise as a primary or secondary airway clearance technique (3) . This study examined daily exercise patterns in a large sample of adolescents with CF using a daily phone diary (DPD) measure.

Method: Data were collected in a large, longitudinal intervention trial to increase adherence and reduce family conflict. Of the 117 families eligible for the study, 103 had DPD data at baseline and were included in the current study. Mean age of the participants was 13.4 years (SD = 1.8), approximately half were female (47%), and mean FEV 1 % predicted was 82.1%.The DPD is a phone-based diary that measures activities, companions, and mood over a 24-hour period using a cued-recall procedure (4) . For all activities lasting longer than five minutes, adolescents reported the type of activity, duration, who was present, and a mood rating. A set of 3 DPDs (2 weekdays, 1 weekend day) was collected for each participant at each assessment. Mood was rated on a five-point Likert scale (1= Very Negative, 3= Neutral, 5= Very positive).Results: A majority of the respondents reported completing some exercise over the three-day period (N= 54, 52.4%), with 137 exercise activities reported. The most frequent activities were biking (N= 23 activities, 16.8% of activities), playing sports at home, like basketball (N= 23, 16.8%), and swimming (N= 14, 10.2%). The duration of these activities was generally long, but there was a great deal of variability (mean= 51.9 minutes, SD= 48.5). Mood ratings during exercise were quite positive (mean= 4.07, SD= 0.56, range= 2 to 5). Of those teens who exercised, only 24.1% reported doing so alone, with 35.9% exercising with family members and 20.4% exercising with friends. No difference was found in FEV 1 % predicted between those who did and did not exercise (FEV 1 = 85.0% predicted vs. 81.9%; t= 0.31, p> .05).Discussion: A majority of adolescents with CF reported engaging in exercise activities as part of their daily routine. Exercise typically consisted of informal physical activities at home rather than organized sports, and often included family members and peers. Statistically significant differences in pulmonary functioning between teens who exercised and teens who did not exercise were not found. Future directions include examining prospective, rather than correlational associations between exercise and Grossoehme, D.H. 1, 3 ; Cotton, S. 2 ; Ragsdale, J. 3 ; Seid, M. 4, 1 1. Pulmonary, CCHMC, Cincinnati, OH, USA; 2. Family Medicine, UCCOM, Cincinnati, OH, USA; 3. Pastoral Care, CCHMC, Cincinnati, OH, USA; 4. Health Policy & Clinical Effectiveness, CCHMC, Cincinnati, OH, USA Objective: The first year after a cystic fibrosis (CF) diagnosis is a crisis period for families. Religion/Spirituality (R/S) is important to many Americans, but few studies have examined R/S coping in CF. Developing interventions designed to address R/S coping is challenging. The purpose of this study was to develop a model of how R/S is used by parents to cope during the first year following the diagnosis of their child with CF.

Methods: This qualitative Grounded Theory study used semi-structured interviews of parents of children diagnosed with CF in the past year. A theoretical model of parental use of R/S during this period was created, centered on a "core dimension" (the most fruitful explanation for the social process being examined) and subsidiary, "primary dimensions" (the means by which the core dimension proceeds).Results: Fifteen children (3-17 mos) were diagnosed with CF during this period. Of all parents approached, 88% agreed to participate; 15 interviews were completed; 9 were mothers. All self-identified as Caucasian; 6 were Catholic, 5 Baptist, 4 Christian. The core dimension is that R/S helps parents make meaning. One primary dimension by which this occurs is viewing God as a personal, active and intervening God ("We tried for year and a half to have another child and we couldn't. My husband started to think there was something wrong. I said, 'No, when God wants us to get pregnant, He'll make sure we get pregnant. Then my daughter was diagnosed; I said, 'I know why now we didn't get pregnant. I didn't wanna be pregnant but God said, "No, you're gonna have another one…obviously, He had His plan and I had mine and He just overruled my plan"). A 2nd is feeling supported by God ("…moments I feel so helpless but I know that He knows what that feels like because He had to watch Jesus, you know. So I try to also put that in perspective that He understands. I'm very thankful He trusted me enough to give N. to me. She's a lot of hard work; it's awesome to be loved enough by God to be honored to have her. Everything we've been through, we haven't gone through alone; God has been there. He has given strength to us. It's not Him that's doing this to her; it's through Him that we're getting through this"). Finally, by finding hope ("I know God doesn't give me more than I can handle; God has more confidence in me than I have in myself"). More parents reported religious beliefs relating to decision-making or adherence than not (e.g., "the whole g-tube thing for me was huge…we used prayer instead and thanks to God she doesn't need one" and "faith…helps me do things I need to do that aren't easy, like the wound care …it's hard to hurt her but…God gives me strength to go through with it."). Stories describing a relationship with God/Higher Power appear to provide the most significant meaning, versus religiosity based on received instruction from one's faith group.Conclusions: R/S is an important part of making meaning, providing hope, and contributing to decision-making and adherence. CF teams should consider chaplains a standard element of care offered. This model can be the basis of psychosocial interventions, including those aimed at improving adherence.

Sintnicolaas, C. 1 ; Peters, J. 2 ; Yntema, J. 1 1. Paediatrics; Respiratory Medicine, Radboud University Medical Centre, Nijmegen, Netherlands; 2. Department of Medical Psychology, Radboud University Medical Centre, Nijmegen, Netherlands Background: Optimizing Quality of Life (QoL) is considered an important goal in the management of patients with a chronic disease. The cystic fibrosis (CF) specific questionnaire (CFQ) has been developed as a tool to monitor QoL in CF patients.Admissions for intravenous antibiotics are an important part of the maintenance of lung function in cystic fibrosis (CF). Admissions involve a multidisciplinary team and patients typically interact with many "microsystems" within the inpatient setting, including radiology, interventional radiology, pharmacy, and operating rooms which can impact the inpatient experience. Inpatient therapy often lasts 1-3 weeks, and the success of therapy is generally based on objective measures such as lung function. Although useful, objective measures alone fail to capture many aspects of the inpatient experience important to patients and families. The Healthcare Matrix is a tool that was developed to link the Institute for Medicine aims for improvement in healthcare and the six ACGME Core Competencies for healthcare providers (1, 2) . It provides a conceptual framework that projects an episode of care as an interaction between quality outcomes and the skills, knowledge, and attitudes of healthcare providers necessary to effect those outcomes. Previous use of the matrix has primarily depended upon healthcare provider assessment of episodes of care. The purpose of this study was to assess whether the Healthcare Matrix can be used as a tool to allow CF patients and families to evaluate the inpatient hospitalization, combining it with typical objective measures of improvement such as improved lung function.Methods: CF patients admitted to the hospital for pulmonary exacerbations had measurements of lung function, oximetry, and weight done on admission and at twice weekly intervals. A subjective analog score of health was completed on admission and weekly. Completion of a Healthcare Matrix questionnaire was done by parents (and patients) and discussed at a weekly care conference with the attending physician, residents, and multidisciplinary team present. The Matrix as completed by the parent/patient (score: 0 or 1) assessed whether the hospital stay was: 1) Safe; 2) Timely; 3) Efficient 4) Equitable; and 5) Patient-and family-centered. Effectiveness was assessed using improvements in lung function, weight, and oxygen levels as well as the subjective improvement score. The health care team was judged for 1) Overall care; 2) Medical knowledge; 3) Interpersonal and communication skills; 4) Professionalism; and 5) Integrated performance of the inpatient microsystems.Results: The Matrix Score can be used to define parent, patient and caregiver-agreed perfect care as well as providing immediate feedback from patients and families on opportunities for team learning and improvement. Changes in Health Matrix scores over time provide an assessment of how hospital systems are meeting the needs of CF patients and families. Gould, A.L.; McGeachey, A.; Mantell, P.; Riale, S.; Sawicki, G.S. Childrens' Hospital-Boston, Boston, MA, USA Background: Airway clearance treatments (ACT) using chest physical therapy (CPT), high frequency chest compression vest (HFCC), or other modalities, are an essential part of the management of pulmonary exacerbations (PEx) in patients with cystic fibrosis (CF). At Children's Hospital Boston due to staffing and time constraints, hospitalized patients with CF received 3 ACT each day between 8:30 am and 5:00 pm. Although optimal frequency of ACT during treatment for PEx is undetermined, increasing the frequency of these ACT may improve patient outcomes. We therefore implemented a quality improvement (QI) initiative to increase ACT by including an evening treatment using HFCC during hospitalizations for PEx.

Methods: Beginning in November 2008 all patients >6 yrs of age admitted for PEx were offered an evening HFCC treatment utilizing Hill-Rom's Vest Airway Clearance System. The HFCC equipment is managed and cleaned by nurses and the treatment is completed by the patient and caregiver following education and demonstration of proper device use by physical therapists. The nursing staff documents the time of the HFCC treatment. As part of our QI initiative, a nursing staff survey assessed nurses' perceptions of the benefits of the additional HFCC treatment and its effect on nursing workload. A patient/parent satisfaction questionnaire (PPQ) on ACT was also administered during hospitalizations before and after implementation of the evening HFCC program.Results: In the four months prior to the initiation of the evening HFCC treatment, 63 hospitalized patients received an average of 2.3 treatments per day. On 49% of hospitalized days they received 3 or more treatments. Prior to the QI project, 13 patients completed the PPQ and 69% were very or extremely satisfied with the ACT received during their admission. From November 2008-April 2009, 37/46 (80%) eligible patients participated in the evening HFCC program. The average number of treatments each day increased to 3.0, and on 73% of hospitalized days patients received 3 or more treatments. After the initiation of the evening HFCC program, 13 PPQ were completed; 85% were very or extremely satisfied with the ACT received during the admission and 85% felt the evening HFCC treatment was beneficial or very beneficial. Many patients chose to do the evening treatment after 9:30 pm; HFCC treatments allowed patients more flexibility of treatment time. Nursing surveys (n=35) revealed that although 55% felt the evening HFCC treatment put an increased burden on the nursing staff, 60% felt it improved patients' ability to clear secretions, and 70% felt it improved patient/parent satisfaction. The primary expense of this program is the cost of disposable vests, which averaged $2025/month, significantly less than the cost of physical therapists extending their shift through 9:00 pm which would be over $5000/month.Conclusion: This QI initiative was successful in increasing the number of ACT patients with CF hospitalized for PEx received each day and appears to be a cost-effective way to increase daily ACT frequency. Patients, parents, and nursing staff reported high satisfaction with the addition of an evening HFCC treatment.This project was partially funded by a 2009 CFF QI grant. Research has shown that good nutritional status is associated with better lung function and pulmonary outcomes in people with CF [1] . However, many patients do not meet their calorie intake and weight gain goals [2] . We have observed that many patients need additional education Background: Medical professionals caring for patients with chronic illness, such as cystic fibrosis (CF), struggle to maintain appropriate boundaries that allow for a safe relationship with a patient based on the patient's needs. Maintaining professional boundaries is difficult due to the nature of support provided, as the patient requires increasing degrees of care. When a professional violates a boundary the indiscretion may exploit the patient's vulnerable position.

Recently, we observed an increase in boundary violations among licensed professionals and unlicensed employees directly involved with the care of inpatients with CF. Following discussions with nursing leadership, a plan was developed to address professional boundaries and assess the staff's beliefs related to boundaries and the impact of education on these beliefs. We aimed to address the needs identified by hospital staff for setting appropriate professional boundaries with patients by developing and conducting educational sessions with the staff.Methods: Two of the authors (KB, LG) led interactive educational sessions with hospital staff about appropriate professional boundaries. The objectives of the sessions were: defining professional boundaries; differentiating between boundary crossings and boundary violations; and techniques for managing boundary violations. A written clinical boundaries inventory (CBI) was administered to the participants prior to and following the sessions. The CBI is a thirty-question tool ranking the acceptability of behaviors on a 3 point Likert scale: 1) Never ok, 2) Sometimes ok, or 3) Always ok. To assess the professional's level of awareness, 4 clinical scenarios were also verbally processed as acceptable behavior, potential boundary crossing, or boundary violation.Results: To date, 40 staff have participated in the educational sessions. CBI obtained after the intervention showed that 81% of respondents changed one or more answers, and 42% of these participants changed 7 or more of their answers. Of those responses changed, 63% of the answers were to a more appropriate boundary. Situations that were frequently answered inappropriately after the sessions included contacting a patient while not working and offering one's family as a socialization option for a patient/patient's siblings. When verbally processing the professional's level of awareness, participants responded appropriately to the scenarios presented. Three scenarios were routinely assessed as questionable boundary crossings: co-ed professional/patient relationships, relationships with former patients who are no longer in the care of the professional, and agreement to act as an employment reference for a patient/patient's family member.Discussion: We conclude that professional boundaries can be improved following an educational intervention. Future sessions will involve additional healthcare professionals, including physicians. We plan to re-administer the CBI within 6 months of the initial training to assess the long-term effects the training had on the beliefs of the participants. We anticipate that ongoing education may be needed. A potent inhaled agent, Equanox is a 50:50 mix of nitrous oxide and oxygen. It provides analgesia, some sedation and relieves anxiety before and during painful procedures. Prior to this an increasing number of children with cystic fibrosis (CF) were requiring general anaesthetic for insertion of long lines. Some requiring gripper needle insertion were becoming increasingly distressed during the procedure. With the assistance of the Paediatric Pain team, it was established that Equanox therapy could be beneficial to this client group. It was observed that Equanox administration was optimised when used in conjunction with input from the play specialist.

Method: Patient satisfaction data was collected over the year April 2008 to March 2009 by means of a simple tick box questionnaire after each procedure using Equanox. Equanox therapy was used for line and gripper needle insertion in CF children on 22 occasions throughout this period. Eight individuals, six boys and two girls with an age range of 4 to 15 years received Equanox therapy during this period. Patients were asked to rate how satisfied they were with the procedure using 4 categories: Poor, Good, Very Good and Excellent.Results: Results showed that on 50% of the occasions that Equanox was used, patients scored their satisfaction as Excellent. On two occasions, patients scored their satisfaction as Very Good, while on 5 occasions scored Good. This means that for over 80% of the occasions on which Equanox was used, children felt it was a positive outcome. No-one scored the procedure as Poor but on two occasions, two individuals became very distressed and could not co-operate with the procedure. One patient felt happy to have his line inserted without the use of Equanox as he felt he no longer required it.Conclusion: Nurse-led Equanox therapy negates the need for some children to be admitted to day case units for line insertion under general anaesthetic. Many studies show it to be a safe and effective analgesic when used for procedure related pain. Its success rate within the CF population at RHSC, although small has encouraged the team to use it more often and undertake training to allow members of the CF team to administer this therapy. Infection control measures currently utilized in the clinic include: hand hygiene prior to and after any interaction with the patient; all staff don gown and gloves prior to entering exam room; the waiting room is arranged so that families may distance themselves minimally three feet from other families; no shared toys or magazines in waiting room; clinic appointment times are arranged to minimize wait time in the waiting room.

Parents of children with CF were asked to complete a seven question survey during their child's clinic visit. Forty-eight parents completed the survey. The surveys were collected ensuring anonymity.Three questions focused on parent knowledge of infection: 1) preventing infection as an important part of their child's care 2) signs of lung infection; and 3) their knowledge that lung infections can cause lung damage.Three questions addressed parent perception of infection control measures utilized in the clinic: 1) are all staff entering the patient exam room garbed in a gown and gloves; 2) utilization of gown and gloves by all staff decreases the risk of infection exposure to their child; and 3) the size of the waiting room allows families the ability to avoid close contact with other CF patients.One question inquired whether parents found the information mailed after clinic visit, discussing infection and the spread of germs, as helpful.The survey revealed that 98% of parents strongly agree or agree that preventing infections is an important part of their child's care. Most (96%) of parents identified change in cough, change in mucous coughed up, fever, and change in appetite as signs of possible lung infection. All (100%) parents indicated that lung infections can cause lung damage.Regarding infection control measures in the clinic, 96% of parents surveyed indicated that staff don gown and gloves prior to coming into the exam room. Eighty-one percent of parents indicated that their child was at decreased risk of germ exposure because staff donned gown and gloves prior to entering the exam room. Seventy-three percent of parents surveyed indicated that they felt the waiting room was large enough to decrease exposure to germs while waiting for their appointment.Ninety-eight percent of parents surveyed indicated that the information sent after their child's appointment about infection and spread of germs is helpful.

Strandner, K. 1 ; Herlitz, K. 1 ; Gilljam, M. 2 ; Lindblad, A. 1 1. Dept of Pediatrics, Queen Silvia Childrens Hospital, Gothenburg, Sweden; 2. Respiratory Medicine and Allergology, Sahlgrenska University Hospital, Gothenburg, Sweden Background: At the combined adult and paediatric West Swedish Cystic Fibrosis (CF) Centre patients are seen every 4-6 weeks. Treatment with IV antibiotics for ten days is started early if there is a clinical suspicion of pulmonary exacerbation and most patients administer the therapy at home without support from homecare teams but can contact the centre when needed. Few patients are treated continuously with inhaled antibiotics.Aim:To investigate to what degree the Fuchs criteria of infectious exacerbation were fulfilled at start of intravenous antibiotic therapy and impact on work and school attendance.Methods: All patients who started IV antibiotics during Sept 1-Dec 31, 2008 were included. Symptoms and signs of exacerbation were recorded. The number of peripheral venous catheters (pvc) needed, number of days on IV, days in hospital and days on sick leave from school/work were noted.Results: Forty patients (20 females) with a median age of 23 years (range 6-66) started a total of 47 IV treatments. Seven patients started 2 treatments. FEV1 at start was 64±20% and 25 (62%) were chronically infected with P. aeruginosa. Twenty patients (50%) had a totally implantable venous access system while 20 patients (50%) used pvc. At discontinuation patients were seen in clinic (36 patients) or had a telephone consultation with the nurse (11 patients).The treatment was extended up to 15 days (range 11-15) in seven patients. Two patients were treated in hospital for 9 and 4 days respectively, both for eradication therapy due to first isolation of P. aeruginosa. At IV start 10/47 (21%) fulfilled Fuchs criteria. Five of 15 children (33%) were absent from school/day-care for median 9 (range 2-10) days. Of 31 treatments in adults, absence from work/studies was recorded for 16 (52%) for a median of 9 (range 3-10) days. The median number of pvc used was 2/patient (range 1-6). At follow-up FEV1 improved with 5.7±6.4% (p<0.001), while the weight was unchanged (36 treatments).Conclusion: Less than 25% fulfilled the Fuchs criteria for pulmonary exacerbation at start of intravenous antibiotic therapy. Hospitalization was seldom necessary and the majority of the patients continued their work or studies as usual.

Müller, F.C. Dept. Pediatrics III, University Heidelberg, Pediatric Pulmonology, Cystic Fibrosis Centre & Infectious Diseases, Heidelberg, Germany Objective: Teenagers with cystic fibrosis (CF) are now in a much better health condition compared to 20 years ago, and therefore they stay less frequently in the hospital. As a consequence, there is less time to provide the teenagers comprehensive information about the disease and treatment opportunities. Teaching in small groups, such as for asthma and atopic dermatitis patients, are of limited use for teenagers with CF, as many of them are already colonized with resistant pathogens and can therefore not attend. Teenagers like to obtain information from the internet and to visit chat rooms. Therefore, in addition to the information provided by the CF team in the outpatient department and during hospital admissions, we developed a CD and internet forum "mukoteens" especially designed for teenagers: 1) more information about the disease and treatment opportunities, where the teenagers can decide when to obtain the information and how specific; 2) for a better therapy adherence and compliance in this particular age group for a better quality of life and life expectancy; and 3) for improved communication about the disease between patients of the same age group.Methods: Adolescent CF patients of the cystic fibrosis center Heidelberg were asked, what topics are of interest for a better knowledge about the disease and treatment strategies. The cystic fibrosis team of the centre and in particular the MDs, a psychiatrist, a gynecologist, a physical therapist, a social worker and a dietician did further contribute. Topics of relevance were selected by the team, questions were collected and cartoons, animations, and videos were generated together with a female and male teenager patient from the center in cooperation with a professional media designer.Results: A CD and a corresponding internet platform "mukoteens" were developed for the following topics: respiration, alimentation, development, pharmacology of drugs, physiotherapy exercises as well as for partnership, sexuality, and desire to have children. The two teenagers generated a diary of their patient history from diagnosis of the disease to their problems to find a apprenticeship position for them. So far, between 4000 and 7000 visits of the mukoteens internet forum were recorded per month.Conclusions: In addition to the information provided by the CF team to teenage patients in the outpatient department and in the hospital about the disease and the treatment opportunities, a CD and internet forum can provide self-regulated information, and can improve the communication between the patients for a better compliance and quality of life.Supported by an unrestricted research grant by Asche Chiesi GmbH, Hamburg, Germany. Background: Our cystic fibrosis (CF) Program is provided through a small multi-disciplinary clinic for both pediatric and adult care. Transfer to adult clinic usually occurs at age 18. With the exception of the physician, members of the CF team provide care throughout the lifespan, following patients at both pediatric and adult clinics which are held at the same setting.

Objective: Our objective was to implement an effective education/transition program to meet the needs of our combined pediatric/adult CF program.Methods: A literature review was conducted to explore strategies used for the successful transition of persons from pediatric to adult care in CF clinics, and the meaning of "transition". Data were collected and reviewed regarding established programs from the literature, NACFC conferences and the Canadian CF Nurses Interest Group, as well as chronic illness models, developmental/theoretical principles, health promotion framework, education techniques for lifelong learning, Standards of Care and Clinical Practice Guidelines for CF. Our CF team held meetings since Dec 2007 to develop an education/transition program to meet the needs of our clinic, using our clinic population statistics to guide priority planning.Results: The literature review revealed little practical advice relevant to our setting. Definitions of "transition" varied. The few formal transition programs in Canada or published internationally did not fully meet the needs of our clinic population. Most programs focused on the processes involved with transfer to a separate adult care facility, and did not describe a formal education program for combined CF pediatric/adult settings. Given these findings, our team decided it would be necessary to develop a unique program for our setting. We developed a definition of "transition" and created a lifespan CF Education/Transition Program Framework. This framework addresses patients' education needs from the time of diagnosis to end of life care, and incorporates an adolescent care pathway to facilitate transition to adult care. It includes learning needs assessments, evidence-based multi-disciplinary clinical documentation tools, and patient/family educational resources. Since 53% of clinic patients were adolescents and/or young adults, this group was targeted as a priority for program development. A CF team documentation tool was drafted for use with the "Adolescent Care Pathway" and early adult care, as well as a "CF Readiness to Transfer to Adult CF Clinic Questionnaire," Adolescent Transition pamphlets, and other resources.Conclusions: Models of transition are still evolving and there are few data on program efficacy. The development of our CF clinic's formal lifespan Education/Transition Program will be the first of its kind in Canada, and was developed to meet the specific patient program needs at our clinical setting. We aim to pilot the adolescent and early adult pathways of our program and incorporate an ongoing evaluation process. Methods: A 60-item questionnaire assessing CF knowledge in five areas (general complications, lung topics, gastrointestinal topics, reproduction, and genetics) was developed and validated. One hundred CF patients (median age: 26.0 years, range 17-49 years; median FEV1: 57.0% predicted, range 20-127% predicted) completed the questionnaire. Level of CF knowledge was correlated with clinical and sociodemographic characteristics.

Results: Validation of the questionnaire showed acceptable internal consistency (α = 0.75) and high test-retest reliability for all domains (0.76-0.97). Overall CF knowledge scores showed that patients had a good understanding of the disease (mean score = 72.4%, SD = 13.1), with greater knowledge of lung and gastrointestinal topics (mean score = 81.6%, SD = 11.6) than reproduction and genetics as they relate to CF (mean score = 57.9%, SD = 24.1). Female gender and receiving post-secondary (post-high school) education were significantly correlated with higher CF knowledge scores in all domains (p < 0.05). There was no correlation between CF knowledge scores and lung function, nutritional status, or socioeconomic status.Conclusions: This study validated a questionnaire that can be utilized to assess CF knowledge in adult patients. Our results indicate that CF patients have a good understanding of most aspects of their disease, although clear deficiencies, particularly regarding reproduction and genetics as they relate to CF, are common. These deficiencies should be considered when developing a transition or adult CF education program.

Dellon, E.P. 1 ; Sawicki, G.S. 3 ; Wolfe, J. 3 ; Hanson, L.C. 2 1. Pediatrics, University of North Carolina, Chapel Hill, NC, USA; 2. Medicine, University of North Carolina, Chapel Hill, NC, USA; 3. Medicine, Children's Hospital Boston, Boston, MA, USA Introduction: Many patients with advanced CF lung disease receive intensive treatments such as non-invasive and invasive mechanical ventilation which are intended to sustain life. There are no standards for the timing or content of patient-physician discussions about preferences for the use of these treatments. We aimed to describe physician practices for discussing preferences for the use of intensive treatments with patients with advanced CF lung disease, including timing and content as well as barriers and facilitators to discussions.Methods: Physicians at 2 large CF care centers completed an internetbased survey about their current practice for discussing intensive treatment preferences with CF patients. Summary statistics were used to analyze responses. Physicians were also interviewed about their recommendations for improving treatment preference discussions, and recurrent themes were identified using qualitative content analysis.Results: Surveys have been completed by 23 of 35 (66%) eligible physicians in this ongoing study, and interviews by 17 (49%). Physicians report a mean of 16 years experience in CF care, with half of physicians providing primary CF care to 20 or more patients. Most (96%) felt the primary CF physician should initiate these discussions with their patients. Opinions were divided on whether first discussions should take at a time of medical stability (48%) versus during acute illness or a major decline in health (52%). Wide variation in the content of discussions was reported, with only 3 issues always being included: potential risks and benefits of intensive treatments, lung transplant preference, and disease prognosis. Major barriers to discussions identified were overly optimistic expectations about prognosis and about effectiveness of intensive treatments on the part of patients/families, concern for taking away hope, prognostic uncertainty, and competing demands for physicians' time. Major facilitators of discussions included lung disease severity, transplant status, and patient/family inquiry about intensive treatments. Themes emerging from interviews included lack of training in discussing treatment preferences and difficulty identifying the appropriate time to initiate discussions due to variations in disease trajectory and in patient/family readiness. All physicians endorsed initiating discussions when patients are stable enough to participate and felt that educational tools for patients would facilitate discussions and informed decision making about the use of intensive treatments.Conclusions: CF physicians feel responsible for initiating discussions about intensive treatment preferences with their patients, but identify lack of formal training, difficulty determining the appropriate time to initiate discussions, and other patient and physician factors which are barriers to these discussions. Consensus about timing and content of discussions and also development and use of educational tools may facilitate discussions and decision making about intensive treatments.Supported by the National Palliative Care Research Center.